+ All Categories
Home > Documents > BIOCHEMISTRY Copyright © 2020 14-3-3 binding creates a ... · 9/1/2020  · Center Göttingen,...

BIOCHEMISTRY Copyright © 2020 14-3-3 binding creates a ... · 9/1/2020  · Center Göttingen,...

Date post: 08-Sep-2020
Category:
Upload: others
View: 0 times
Download: 0 times
Share this document with a friend
17
Menzel et al., Sci. Signal. 13, eaaz1436 (2020) 1 September 2020 SCIENCE SIGNALING | RESEARCH ARTICLE 1 of 16 BIOCHEMISTRY 14-3-3 binding creates a memory of kinase action by stabilizing the modified state of phospholamban Julia Menzel 1 , Daniel Kownatzki-Danger 2 *, Sergiy Tokar 3 *, Alice Ballone 4,5 , Kirsten Unthan-Fechner 1 , Markus Kilisch 1 , Christof Lenz 6,7 , Henning Urlaub 6,7 , Mattia Mori 4 , Christian Ottmann 5 , Michael J. Shattock 3 , Stephan E. Lehnart 2,8 , Blanche Schwappach 1,7,8† The cardiac membrane protein phospholamban (PLN) is targeted by protein kinase A (PKA) at Ser 16 and by Ca 2+ / calmodulin-dependent protein kinase II (CaMKII) at Thr 17 . -Adrenergic stimulation and PKA-dependent phos- phorylation of Ser 16 acutely stimulate the sarcoplasmic reticulum calcium pump (SERCA) by relieving its inhibition by PLN. CaMKII-dependent phosphorylation may lead to longer-lasting SERCA stimulation and may sustain maladaptive Ca 2+ handling. Here, we demonstrated that phosphorylation at either Ser 16 or Thr 17 converted PLN into a target for the phosphoadaptor protein 14-3-3 with different affinities. 14-3-3 proteins were localized within nanometers of PLN and endogenous 14-3-3 coimmunoprecipitated with pentameric PLN from cardiac membranes. Molecular dynamics simulations predicted different molecular contacts for peptides phosphorylated at Ser 16 or Thr 17 with the binding groove of 14-3-3, resulting in varied binding affinities. 14-3-3 binding protected either PLN phosphosite from dephosphorylation. -Adrenergic stimulation of isolated adult cardiomyocytes resulted in the membrane recruitment of endogenous 14-3-3. The exogenous addition of 14-3-3 to -adrenergic–stimulated cardiomyocytes led to prolonged SERCA activation, presumably because 14-3-3 protected PLN pentamers from dephosphorylation. Phosphorylation of Ser 16 was disrupted by the cardiomyopathy-associated ∆Arg 14 mutation, implying that phosphorylation of Thr 17 by CaMKII may become crucial for 14-3-3 recruitment to ∆Arg 14 PLN. Consistent with PLN acting as a dynamic hub in the control of Ca 2+ handling, our results identify 14-3-3 binding to PLN as a contractility-augmenting mechanism. INTRODUCTION The cardiac excitation-contraction cycle relies critically on the spatiotemporal control of intracellular Ca 2+ release and sequestra- tion in the sarcoplasmic reticulum (SR) (1). In cardiomyocytes, the ryanodine receptor (RyR2) Ca 2+ release channel and the Ca 2+ pump SERCA cooperate in large protein assemblies to ensure the physio- logical plasticity of Ca 2+ handling (2). The small membrane protein phospholamban (PLN) negatively regulates the pumping activity of SERCA and thereby affects the reuptake kinetics of Ca 2+ into the SR. The protein is dispensable in mice, perhaps because it regulates only 40% of the SERCA pumps (34). Yet, mutations found hetero- zygously in patients with inherited dilated cardiomyopathy (DCM) suggest that humans require the full complement of PLN (57). Protein kinase A (PKA)–dependent phosphorylation targets PLN on Ser 16 , relieves the inhibitory effect of PLN on SERCA, and hence stimulates SERCA (89). During -adrenergic stimulation, this mechanism results in increased force generation and an accelerated relaxation of cardiac muscle. The adjacent Thr 17 is targeted by Ca 2+ -calmodulin–dependent kinase (CaMKII) in response to various stimuli (1011), although phosphorylation of Ser 16 is sufficient to mediate the effects of -adrenergic signal transduction (12). Aside from its regulation by phosphorylation, PLN exists in an equilibrium between two assembly states, a monomeric and a pentameric form, which can be observed as an SDS-resistant pentamer (91314), consistent with the high stability of the oligomer observed by fluo- rescence resonance energy transfer (FRET) in living cells (15). This equilibrium and a spectrum of conformations assumed by the cyto- plasmic domain of PLN are central to the regulatory effect of Ser 16 or Thr 17 phosphorylation on PLN (1315) and may hence be targeted by other modulatory mechanisms including protein-protein inter- actions (16). The regulatory subunit of protein phosphatase 1c (PP1c), PPP1R3A (217), recruits not only the phosphatase but also two additional proteins that inhibit PP1c, inhibitor-1 (I-1) and the small heat shock protein Hsp20. In previous work on cardiac adenosine triphosphate (ATP)– sensitive potassium channels (K ATP ), we dissected arginine (Arg)– based trafficking motifs in K ATP subunits (18). Arg-based motifs bind the vesicle coat complex COPI and in many cases also 14-3-3 phosphoadaptor proteins. 14-3-3 proteins are encoded by a family of seven related genes, which historically have been called isoforms (1921). They bind to various client proteins by engaging with target sequences that often but do not always contain phospho- residues. Binding may stabilize specific conformations of a client protein or sterically block alternative protein-protein interactions. As a result, the subcellular localization, stability, or posttranslational modification state of the client protein may be affected by 14-3-3 binding. Client binding involves a highly conserved ligand-binding 1 Department of Molecular Biology, Universitätsmedizin Göttingen, Humboldtallee 23, 37073 Göttingen, Germany. 2 Heart Research Center Göttingen, Department of Cardiology & Pneumology, Universitätsmedizin Göttingen, Robert-Koch-Straße 42a, 37075 Göttingen, Germany. 3 School of Cardiovascular Medicine and Sciences, King’s College London, Westminster Bridge Road, London SE17H, UK. 4 Department of Biotechnology, Chemistry and Pharmacy, Department of Excellence 2018-2022, University of Siena, Via Aldo Moro 2, 53100 Siena, Italy. 5 Laboratory of Chemical Biology, Department of Biomedical Engineering and Institute for Complex Molecular Systems, Eindhoven University of Technology, P. O. Box 513, 5600MB Eindhoven, Netherlands. 6 Bioanalytics Group, Institute of Clinical Chemistry, University Medical Center Göttingen, Robert-Koch-Straße 40, 37075 Göttingen, Germany 7 Max-Planck Institute for Biophysical Chemistry, Am Faßberg 11, 37077 Göttingen, Germany. 8 Cluster of Excellence “Multiscale Bioimaging: From Molecular Machines to Networks of Excitable Cells” (MBExC), University of Goettingen, Robert-Koch-Straße 40, 37075 Göttingen, Germany. *These authors contributed equally to this work. †Corresponding author. Email: [email protected] Copyright © 2020 The Authors, some rights reserved; exclusive licensee American Association for the Advancement of Science. No claim to original U.S. Government Works on December 28, 2020 http://stke.sciencemag.org/ Downloaded from
Transcript
Page 1: BIOCHEMISTRY Copyright © 2020 14-3-3 binding creates a ... · 9/1/2020  · Center Göttingen, Robert-Koch-Straße 40, 37075 Göttingen, Germany 7Max-Planck Institute for Biophysical

Menzel et al., Sci. Signal. 13, eaaz1436 (2020) 1 September 2020

S C I E N C E S I G N A L I N G | R E S E A R C H A R T I C L E

1 of 16

B I O C H E M I S T R Y

14-3-3 binding creates a memory of kinase action by stabilizing the modified state of phospholambanJulia Menzel1, Daniel Kownatzki-Danger2*, Sergiy Tokar3*, Alice Ballone4,5, Kirsten Unthan-Fechner1, Markus Kilisch1, Christof Lenz6,7, Henning Urlaub6,7, Mattia Mori4, Christian Ottmann5, Michael J. Shattock3, Stephan E. Lehnart2,8, Blanche Schwappach1,7,8†

The cardiac membrane protein phospholamban (PLN) is targeted by protein kinase A (PKA) at Ser16 and by Ca2+/calmodulin-dependent protein kinase II (CaMKII) at Thr17. -Adrenergic stimulation and PKA-dependent phos-phorylation of Ser16 acutely stimulate the sarcoplasmic reticulum calcium pump (SERCA) by relieving its inhibition by PLN. CaMKII-dependent phosphorylation may lead to longer-lasting SERCA stimulation and may sustain maladaptive Ca2+ handling. Here, we demonstrated that phosphorylation at either Ser16 or Thr17 converted PLN into a target for the phosphoadaptor protein 14-3-3 with different affinities. 14-3-3 proteins were localized within nanometers of PLN and endogenous 14-3-3 coimmunoprecipitated with pentameric PLN from cardiac membranes. Molecular dynamics simulations predicted different molecular contacts for peptides phosphorylated at Ser16 or Thr17 with the binding groove of 14-3-3, resulting in varied binding affinities. 14-3-3 binding protected either PLN phosphosite from dephosphorylation. -Adrenergic stimulation of isolated adult cardiomyocytes resulted in the membrane recruitment of endogenous 14-3-3. The exogenous addition of 14-3-3 to -adrenergic–stimulated cardiomyocytes led to prolonged SERCA activation, presumably because 14-3-3 protected PLN pentamers from dephosphorylation. Phosphorylation of Ser16 was disrupted by the cardiomyopathy-associated ∆Arg14 mutation, implying that phosphorylation of Thr17 by CaMKII may become crucial for 14-3-3 recruitment to ∆Arg14 PLN. Consistent with PLN acting as a dynamic hub in the control of Ca2+ handling, our results identify 14-3-3 binding to PLN as a contractility-augmenting mechanism.

INTRODUCTIONThe cardiac excitation-contraction cycle relies critically on the spatiotemporal control of intracellular Ca2+ release and sequestra-tion in the sarcoplasmic reticulum (SR) (1). In cardiomyocytes, the ryanodine receptor (RyR2) Ca2+ release channel and the Ca2+ pump SERCA cooperate in large protein assemblies to ensure the physio-logical plasticity of Ca2+ handling (2). The small membrane protein phospholamban (PLN) negatively regulates the pumping activity of SERCA and thereby affects the reuptake kinetics of Ca2+ into the SR. The protein is dispensable in mice, perhaps because it regulates only 40% of the SERCA pumps (3, 4). Yet, mutations found hetero-zygously in patients with inherited dilated cardiomyopathy (DCM) suggest that humans require the full complement of PLN (5–7). Protein kinase A (PKA)–dependent phosphorylation targets PLN on Ser16, relieves the inhibitory effect of PLN on SERCA, and hence stimulates SERCA (8, 9). During -adrenergic stimulation, this

mechanism results in increased force generation and an accelerated relaxation of cardiac muscle. The adjacent Thr17 is targeted by Ca2+-calmodulin–dependent kinase (CaMKII) in response to various stimuli (10, 11), although phosphorylation of Ser16 is sufficient to mediate the effects of -adrenergic signal transduction (12). Aside from its regulation by phosphorylation, PLN exists in an equilibrium between two assembly states, a monomeric and a pentameric form, which can be observed as an SDS-resistant pentamer (9, 13–14), consistent with the high stability of the oligomer observed by fluo-rescence resonance energy transfer (FRET) in living cells (15). This equilibrium and a spectrum of conformations assumed by the cyto-plasmic domain of PLN are central to the regulatory effect of Ser16 or Thr17 phosphorylation on PLN (13, 15) and may hence be targeted by other modulatory mechanisms including protein-protein inter-actions (16). The regulatory subunit of protein phosphatase 1c (PP1c), PPP1R3A (2, 17), recruits not only the phosphatase but also two additional proteins that inhibit PP1c, inhibitor-1 (I-1) and the small heat shock protein Hsp20.

In previous work on cardiac adenosine triphosphate (ATP)– sensitive potassium channels (KATP), we dissected arginine (Arg)–based trafficking motifs in KATP subunits (18). Arg-based motifs bind the vesicle coat complex COPI and in many cases also 14-3-3 phosphoadaptor proteins. 14-3-3 proteins are encoded by a family of seven related genes, which historically have been called isoforms (19–21). They bind to various client proteins by engaging with target sequences that often but do not always contain phospho-residues. Binding may stabilize specific conformations of a client protein or sterically block alternative protein-protein interactions. As a result, the subcellular localization, stability, or posttranslational modification state of the client protein may be affected by 14-3-3 binding. Client binding involves a highly conserved ligand-binding

1Department of Molecular Biology, Universitätsmedizin Göttingen, Humboldtallee 23, 37073 Göttingen, Germany. 2Heart Research Center Göttingen, Department of Cardiology & Pneumology, Universitätsmedizin Göttingen, Robert-Koch-Straße 42a, 37075 Göttingen, Germany. 3School of Cardiovascular Medicine and Sciences, King’s College London, Westminster Bridge Road, London SE17H, UK. 4Department of Biotechnology, Chemistry and Pharmacy, Department of Excellence 2018-2022, University of Siena, Via Aldo Moro 2, 53100 Siena, Italy. 5Laboratory of Chemical Biology, Department of Biomedical Engineering and Institute for Complex Molecular Systems, Eindhoven University of Technology, P. O. Box 513, 5600MB Eindhoven, Netherlands. 6Bioanalytics Group, Institute of Clinical Chemistry, University Medical Center Göttingen, Robert-Koch-Straße 40, 37075 Göttingen, Germany 7Max-Planck Institute for Biophysical Chemistry, Am Faßberg 11, 37077 Göttingen, Germany. 8Cluster of Excellence “Multiscale Bioimaging: From Molecular Machines to Networks of Excitable Cells” (MBExC), University of Goettingen, Robert-Koch-Straße 40, 37075 Göttingen, Germany.*These authors contributed equally to this work.†Corresponding author. Email: [email protected]

Copyright © 2020 The Authors, some rights reserved; exclusive licensee American Association for the Advancement of Science. No claim to original U.S. Government Works

on Decem

ber 28, 2020http://stke.sciencem

ag.org/D

ownloaded from

Page 2: BIOCHEMISTRY Copyright © 2020 14-3-3 binding creates a ... · 9/1/2020  · Center Göttingen, Robert-Koch-Straße 40, 37075 Göttingen, Germany 7Max-Planck Institute for Biophysical

Menzel et al., Sci. Signal. 13, eaaz1436 (2020) 1 September 2020

S C I E N C E S I G N A L I N G | R E S E A R C H A R T I C L E

2 of 16

groove of 14-3-3 proteins. The relative abundance and biological relevance of the isoforms, which differ most in their N- and C-terminal regions, remains poorly understood.

The molecular interactions of Arg-based motifs with COPI and 14-3-3 are crucial to the proper assembly, subcellular localization, and regulation of membrane proteins that present Arg-based signals (22, 23). For KATP and the two-pore domain potassium channel TASK-1 (24), interactions with COPI and 14-3-3 are regulated by PKA-dependent phosphorylation. On the basis of its N-terminal sequence (Fig. 1A), we predicted that PLN has an Arg-based signal that may be regulated in a similar manner (18) and set out to test whether PKA-phosphorylated PLN could bind to 14-3-3 proteins. We found that two distinct phosphosites, phosphorylated (p) Ser16 and pThr17, engaged 14-3-3 with different affinities. We propose that this differential binding adapts the lifetimes of the respective phosphostate to its specific biological purpose.

RESULTS14-3-3 binds to SDS-resistant PLN pentamersPLN is a major target of PKA in cardiomyocytes (25), and its PKA consensus sequence and a putative 14-3-3–binding site (18) are highly conserved (Fig. 1A). To test whether PKA-phosphorylated pSer16-PLN interacted with 14-3-3, we incubated recombinantly expressed 14-3-3 immobilized on a matrix with solubilized cardiac membranes or cytosolic fractions prepared from mouse ventricular tissue (Fig. 1B). The R18 peptide is a high-affinity 14-3-3 ligand (26, 27) that is routinely used to verify the specific interaction of a protein with 14-3-3. Structural models of R18 bound to 14-3-3 obtained by x-ray crystallography demonstrate that R18 occupies the canonical peptide-binding groove of 14-3-3 and hence blocks other ligands from binding. Silver staining (Fig. 1C) of the eluate revealed the presence of bound proteins from the membrane fraction, which we probed by Western blotting using an antibody recogniz-ing generic PKA-dependent phosphosites (Fig. 1D and fig. S1A). The strongest band detected was consistent with the presence of a PKA-phosphorylated PLN pentamer in the 14-3-3 eluate, which can be observed by SDS–polyacrylamide gel electrophoresis (SDS-PAGE) under denaturing conditions (14), possibility due to its high stability (15). To test whether this band corresponded to PLN, we probed the eluates using specific antibodies against PLN (Fig. 1, E and F, and fig. S1B) and pSer16-PLN (Fig. 1, G and H, and fig. S1, C and D). Because preincubation with R18 reduced PLN binding to background levels (Fig. 1, F and H), this experiment identified SDS-resistant PLN pentamers as 14-3-3 ligands that use the canonical ligand-binding groove. PLN monomers, which were present in the solubilized membranes, were not found in the 14-3-3 eluates (Fig. 1, D, E, and G).

In situ biotinylation reveals that PLN is situated in close proximity to 14-3-3 in cardiomyocytesTo complement this analysis with a method that enables the detec-tion of 14-3-3 in proximity to PLN in situ, we fused an engineered ascorbate peroxidase (V5-APEX2) to the PLN N terminus and ex-pressed the fusion protein in neonatal rat cardiomyocytes (NRCMs; Fig. 2A and fig. S2A) using cytosolic green fluorescent protein (GFP) (fig. S2B) or PLN lacking the cytosolic N-terminal domain (fig. S2C) as negative controls for background binding to the avidin matrix and nonspecific APEX2-mediated labeling, respectively. V5-

APEX2–PLN localized to SR membranes as expected (fig. S2A). Six of seven 14-3-3 proteins in the family were specifically labeled when APEX2 fused to full-length PLN was exposed to biotin-phenol and H2O2 (Fig. 2B and data file S1), including 14-3-3, which was the bait in our 14-3-3 affinity purification (Fig. 1B). PLN itself was strongly enriched in both controls as expected based on autolabeling within V5-APEX2–PLN and on the oligomeric states of PLN. On the other hand, the functionally best characterized interaction of PLN with SERCA pumps (8) could not be detected, presumably because SERCA2a is a highly abundant cardiac protein leading to substantial background in the biotin affinity step of the protocol (fig. S2, B and C) and because the transmembrane segment of PLN may remain in close proximity to SERCA2a in SR membranes (fig. S2C). Similarly, another membrane protein that associates with PLN, the PP1c- regulating protein PPP1R3A (2, 17), could not be detected, although soluble proteins that interact with PPP1R3A, PP1c itself and Hsp20, were biotin-labeled (Fig. 2B and data file S1). These two proteins acted as positive controls for the proximity proteomics approach, and their detection underscored that proximity labeling could be due to indirect interactions within larger protein complexes. Hence, the enrichment of six of seven 14-3-3 isoforms could not be inter-preted beyond proximity because 14-3-3 proteins exist as different heterodimers. Binding to PLN could occur through one specific isoform that engages in different heterodimers. However, the peptide-binding groove of 14-3-3 proteins is highly conserved and PLN may directly engage with different 14-3-3 proteins, as occurs with many other 14-3-3 ligands (28). We conclude that PLN and several endogenous 14-3-3 proteins are present within nanometers of each other in cardiomyocytes and that direct or indirect interactions may explain this proximity.

14-3-3 enriches PLN pentamers more efficiently than PLN monomersOur specific enrichment of SDS-resistant PLN pentamers that contain PKA-phosphorylated PLN by 14-3-3 (Fig. 1G) implies an avidity effect in which the two ligand-binding grooves present in the 14-3-3 dimer support binding of 14-3-3 to PLN with a higher apparent affinity than the binding parameters of one peptide ligand binding to a single binding site would allow (29, 30). To further test this idea, we converted most of the PLN present in solubilized mouse cardiac membrane into the monomeric form by gentle heating (14). In the heated membrane input, the monomer was the predom-inant species. However, only 10% of this monomeric PLN was enriched by 14-3-3, whereas 30% of the pentamer present in the solubilized membranes before heating was bound (Fig. 3, A and B). This result supports the concept that an avidity effect is relevant to the interaction between PLN and 14-3-3.

Next, we tested direct binding of 14-3-3 to PLN in an in vitro system. A recombinant protein consisting of the first 31 N-terminal residues of PLN fused to glutathione S-transferase (GST) did not bind to 14-3-3 despite efficient phosphorylation of Ser16 by puri-fied PKA catalytic subunit in vitro (Fig. 3, C and D). This finding may reflect a low affinity of the interaction between pSer16-PLN and one 14-3-3 ligand-binding groove. However, when we exchanged the neighboring residue Thr17 with an alanine, the enrichment of pSer16-T17A-PLN-GST for 14-3-3 was more efficient (Fig. 3, C and D). These data confirm that pSer16 contributes to a 14-3-3–binding site that is weakened by the neighboring Thr17. The low affinity of 14-3-3 binding to the monomeric pSer16-PLN-GST construct is consistent

on Decem

ber 28, 2020http://stke.sciencem

ag.org/D

ownloaded from

Page 3: BIOCHEMISTRY Copyright © 2020 14-3-3 binding creates a ... · 9/1/2020  · Center Göttingen, Robert-Koch-Straße 40, 37075 Göttingen, Germany 7Max-Planck Institute for Biophysical

Menzel et al., Sci. Signal. 13, eaaz1436 (2020) 1 September 2020

S C I E N C E S I G N A L I N G | R E S E A R C H A R T I C L E

3 of 16

with the presence of pSer16-PLN in the 14-3-3–interacting SDS-resistant pentamers but not the monomers (Fig. 1, G and H) and the observed avidity effect (Fig. 3, A and B).

14-3-3 binds with higher affinity to pThr17-PLN than to pSer16-PLNAfter conversion of PLN into monomers by heating, 10% of PLN monomers and 30% of the SDS-resistant pentamers, which have previously been interpreted as a biochemical correlate of endogenous pentamers (9, 14), were enriched on the 14-3-3 matrix (Fig. 3, A and B). The kinase CaMKII targets Thr17 (8, 9), which is adjacent to Ser16 targeted by PKA. Because specific antibodies detecting pThr17-PLN are available (fig. S1, C and D), we probed solubilized cardiac mem-branes that had been heated or not as well as the respective eluates

from the 14-3-3 matrix (Fig. 3, E and F). This antibody was less efficient at detecting monomers compared to pentamers in the input. However, the pThr17 monomer and pentamer were equally enriched by the 14-3-3 matrix, leading us to infer that monomeric pThr17-PLN could bind to 14-3-3 with higher affinity than pSer16- PLN. We observed robust binding of PLN-GST phosphorylated in vitro with CaMKII to 14-3-3 (Fig. 3, G and H). This result indicates that either phosphosite generates a 14-3-3 target motif but with apparently different affinities when monomeric PLN is offered as a target.

We found that all seven 14-3-3 isoforms bound to pThr17-PLN-GST (fig. S3). Although this result does not reflect the interactions between endogenous cardiomyocyte proteins, it does show that direct binding of 14-3-3 to PLN can occur. We also measured 14-3-3

A

B DC

Tissue

Membrane

14-3-3BeadsR18

H

C52

N1

14-3-3 binding motif

PKA target motif

31

14-3-3

Silver stain

MB

P

−+180130

95

554334

26

72

kDa R18

17

180130

9555433426

17

WB:PKA substrate

kDaM

BP

14-3-3

−+Inpu

t 5%

R18

H.s.M.m.

11 2116

M

P

MB

P

14-3-3−+In

put 5

%

kDa

WB:PLN pSer

3426

17

R18 5

PLN

pS

er

(a.u

.)

01234

M

P

MB

P

14-3-3−+In

put 5

%

kDa

WB:PLN

3426

17

R18

G

FE

Inpu

t 5%

MBP

14-3

-3 +

R18

14-3

-3 0

PLN

(a.

u.)

1

2

3

4

***

Inpu

t 5%

MBP

14-3

-3 +

R18

14-3

-3 16

16

5

67 ***

Fig. 1. 14-3-3 binds PLN pentamers from cardiac membranes. (A) Schematic representation of PLN indicating the PKA consensus site and a putative 14-3-3–binding site (18). (B) Scheme for the 14-3-3 affinity purification protocol. (C) Representative silver-stained gel revealing proteins from the membrane fraction copurifying with 14-3-3 and background according to work flow shown in (B) (n = 3 mice per group). (D) Presence of PKA-phosphorylated proteins in the 14-3-3–bound membrane fraction (n = 3 mice per group). See fig. S1A for the full blot including the cytosol fraction. (E) Immunoblotting of the 14-3-3–bound membrane fraction with a PLN antibody (n = 6 mice per group). See fig. S1B for the full blot including the cytosol fraction. (F) Quantification of experiments shown in (E). (G) Detection with phospho-specific anti-pSer16 reveals the presence of pSer16-PLN in the 14-3-3–bound proteins from the membrane fraction (n = 5 mice per group). (H) Quantification of experiments shown in (G). In (F) and (H), data are presented as box and whisker plots with mean and median. A one-way ANOVA analysis showed significant differences between the MBP, 14-3-3 + R18, and 14-3-3 group in (F) (P ≤ 0.0001) and in (H) (P ≤ 0.0002). A Tukey test revealed significant differences in (F) between 14-3-3+ R18 compared to 14-3-3 (***P = 0.0002) and between MBP compared to 14-3-3 (***P = 0.0003) as well as in (H) between 14-3-3 + R18 compared to 14-3-3 (***P = 0.0009) and MBP compared to 14-3-3 (***P = 0.0003). H.s., Homo sapiens; M.m., Mus musculus; R18, R18 trifluoroacetate; WB, Western blot; M, monomer; P, SDS-resistant pentamer; a.u., arbitrary units.

on Decem

ber 28, 2020http://stke.sciencem

ag.org/D

ownloaded from

Page 4: BIOCHEMISTRY Copyright © 2020 14-3-3 binding creates a ... · 9/1/2020  · Center Göttingen, Robert-Koch-Straße 40, 37075 Göttingen, Germany 7Max-Planck Institute for Biophysical

Menzel et al., Sci. Signal. 13, eaaz1436 (2020) 1 September 2020

S C I E N C E S I G N A L I N G | R E S E A R C H A R T I C L E

4 of 16

binding to immobilized pSer16- or pThr17-PLN-GST by surface plasmon resonance (SPR) (fig. S4A) (24). We determined an equi-librium dissociation constant of 4 M (fig. S4, B and C) for the binding of 14-3-3 to pThr17-PLN-GST. It was difficult to obtain reliable sensograms for PKA-phosphorylated wild type (WT) (fig. S4D). Hence, we cannot provide equilibrium dissociation constants for these binding events but conclude that the respective affinities are substantially lower than those observed for pThr17-PLN.

pSer16 and pThr17 determine the conformation of the PLN/14-3-3 complexPossible differences in the molecular contacts between the distinct PLN phos-phoforms and 14-3-3 were addressed by molecular dynamics (MD) simulations, exploiting the conserved mechanism of 14-3-3 and 14-3-3 for phospho- dependent ligand binding (31). This approach was also justified by our observa-tion that 14-3-3 bound to immobilized pThr17-PLN-GST (fig. S3) and that all 14-3-3 proteins engaged with pThr17-PLN. Unrestrained MD trajectories were generated for 250 ns on each system. For the 14-3-3/pSer16-PLN complex, the all-atom root mean square deviations (RMSDs) (fig. S5A) suggested that con-formational convergence is achieved during MD simulation time, whereas the comparison of the root mean square fluctuation (RMSF) between 14-3-3 residues from the complex and apo 14-3-3 (fig. S5B) showed that residues with the highest fluctuation belonged to loops 64 to 88 and 208 to 219 (fig. S5, B and C). The conformational freedom of residues 118 to 190 that belonged to the 14-3-3 amphipathic groove was constrained by the phosphopeptide. Moreover, by cluster analysis of the phosphopeptide and residues within 6 Å (fig. S5, C and D), we found that the most abundant population of MD frames had a peptide conformation (Fig. 4A) that was comparable to that observed in x-ray crystallographic structures (32). pSer16 is located in the amphipathic groove and establishes hydrogen bonds to Lys49 (2.97 Å), Arg129 (1.81 Å), and Tyr130 (1.86 Å), consistent with a mode II binding motif (33).

Similar results were obtained for the 14-3-3/pThr17-PLN complex (Fig. 4B and fig. S5, E to G). The first cluster featuring a population of 45.6% of MD frames was taken as the most represent-ative model (fig. S5H) for further investi-gation and comparison with the 14-3-3/pSer16-PLN complex. The structure of

both peptides in their respective 14-3-3 complexes does not com-pletely overlap because the first three helices of 14-3-3/pSer16-PLN complex show a more opened conformation compared to those in the 14-3-3/pThr17-PLN complex (Fig. 4C). Similar to pSer16-PLN, pThr17-PLN occupies the canonical central position in the amphip-athic groove (32, 33) and establishes H-bond interactions with Lys49 (3.94 Å), Arg56 (1.83 Å), Arg129 (1.68 Å), and Tyr130 (1.64 Å). The spatially similar placement of pSer16 and pThr17 within the groove

A

B

C

V5 APEX

C

HO

B

O

SILAC- labeled proteins

B

Biotinenrichment

PLN (1–52) PLN

(30–52)H2O2

B

B

HO

B

O

B

V5 APEX

PLN

14-3-3 protein

PP1c Hsp20

BBBB

0.0 0.5 1.0 1.50

2

4

4.5 5.0Log2 APEX2-PLN/APEX2-PLN (1–29)

Log 2

AP

EX

2-P

LN/e

GF

P

Fig. 2. PLN proximity proteomics enriches six of seven 14-3-3 isoforms. (A) Scheme of proximity proteomic strategy: N-terminal fusion of the V5-APEX2 peroxidase to full-length PLN was compared to PLN lacking 29 residues of the N terminus. (B) Enrichment of 14-3-3 isoforms (orange circles), PP1c, and Hsp20 with full-length PLN was identified by proximity proteomics in living NRCMs (see data file S1 for enriched proteins). Enrichment compared to two different negative controls is shown cytosolic GFP to assess background binding to the biotin affinity matrix and N-terminally truncated PLN (1–29) to assess specific binding to the exposed region of the membrane protein (n = 5 independent experiments from three NRCM preparations). In (B), only proteins that positively differed from 0 according to a one-sample t test (P < 0.05) in the ratios for both full-length PLN/eGFP and full-length PLN/PLN 1–29 mutant are shown. 14-3-3 was only significantly enriched over eGFP (P < 0.05) but not over PLN 1–29 mutant (P = 0.07). See fig. S2 (B and C) for the intensity of detected proteins. APEX, ascorbate peroxidase; SILAC, stable isotope labeling by amino acids in cell culture; , 14-3-3 delta; , 14-3-3 eta; , 14-3-3 epsilon; , 14-3-3 gamma; , 14-3-3 sigma; , 14-3-3 tau; , 14-3-3 zeta.

on Decem

ber 28, 2020http://stke.sciencem

ag.org/D

ownloaded from

Page 5: BIOCHEMISTRY Copyright © 2020 14-3-3 binding creates a ... · 9/1/2020  · Center Göttingen, Robert-Koch-Straße 40, 37075 Göttingen, Germany 7Max-Planck Institute for Biophysical

Menzel et al., Sci. Signal. 13, eaaz1436 (2020) 1 September 2020

S C I E N C E S I G N A L I N G | R E S E A R C H A R T I C L E

5 of 16

shifts the conformation of the rest of the PLN peptide. As a result, the remain-ing portion of the pThr17 phosphopep-tide is elongated within the groove, placing Ile18, Glu19, and Met20 of PLN in different molecular environments as revealed by electrostatic rendering of 14-3-3 (Fig. 4, A and B).

At least one arginine residue located at the −4, −3, or −2 position relative to the phosphoresidue is a key feature of the canonical 14-3-3–binding phospho-peptide motif (34). This residue stabilizes the phosphopeptide conformation by forming hydrogen bonds and intra-molecular salt bridge with the phosphate group (35). In pSer16-PLN, this role is taken by Arg14 in the −2 position (Fig. 4A), whereas Arg13 (−3 position) is pointing

C

kDa

Phostag-PAGE

3426

WB: 14-3-3 pan H8

CaMKII Inpu

t 20%

GST

3426

SDS-PAGE

3426

T17A

− +S16

A

− +W

T

− +

D

G

Coo

mas

sie

H

kDa

Phostag-PAGE

3426

WB: 14-3-3 pan H8

PKA Inpu

t 20%

GST

3426

SDS-PAGE

3426

T17A

− +S16

A

− +W

T

− +

Coo

mas

sie

PLNpPLN

PLN

PLNpPLN

PLN

WT

S16A

T17A

75°C

17

3426

kDa

P

M

Inpu

t 10%

− +Elut

ion

− +

WB: PLN

A BPentamerMonomer

PLN

nor

m to

inpu

t

75°C+−Elution

*

0.2

0.4

75°C

17

3426

kDa

P

M

Inpu

t 10%

− +Elut

ion

− +

WB: PLN pThr0

0.2

0.4

0.6

75°C+−Elution

E F

0

14-3

-3/p

Ser

P

LN16

P

LN p

Thr

n

orm

to in

put

1714

-3-3

/pT

hr17

PLN

17

0.6

0.8 PentamerMonomer

WT

S16A

T17A

0

0.5

1.0

1.5

0

0.5

1.0

1.5

2.0

Fig. 3. 14-3-3 preferentially binds PLN pentamer, and pThr17-PLN has a higher 14-3-3–binding affinity than pSer16-PLN. (A) Effect of heating cardiac membranes to convert PLN pentamers to monomers on binding of PLN to 14-3-3 (n = 7 mice per group). (B) Quantification of experi-ments shown in (A). The proportion of the re-spective form in the eluate was determined in relation to the pentamer or monomer signal from the input. (C) Binding assay using recombinant, monomeric PLN-GST and the indicated variants to assess 14-3-3 binding after PKA-dependent phosphorylation in vitro (n = 3 independent experiments). (D) Quantification of experiments shown in (C) using normalized values of 14-3-3 intensity and of the slower migrating band in phostag gels. (E) The experiments shown in (A) were probed for pThr17-PLN on independent Western blots (n = 5 mice per group). (F) Quanti-fication of experiments shown in (E). The proportion of the respective form in the eluate was deter-mined in relation to the pentamer or monomer signal from the input. (G) Binding assay using recombinant, monomeric PLN-GST and the indi-cated variants to assess 14-3-3 binding after CaMKII-dependent phosphorylation of PLN-GST (n = 3 independent experiments). (H) Quantification of experiments shown in (G) using normalized values of the 14-3-3 intensity and of the slower migrating band in the phostag gel. Data in (B), (D), (F), and (H) are presented as box and whisker plots with mean and median. A one-way ANOVA analysis showed significant differences between groups in (B) (P ≤ 0.0001) and in (F) (P ≤ 0.0001), and a Tukey test showed significant differences in (B) between pentamer −75°C and monomer +75°C (*P = 0.0143). The multiple comparisons test showed no significant difference in (F) be-tween pentamer −75°C and monomer +75°C (P = 0.5587). M, monomer; P, SDS-resistant pentamer; pPLN, phosphorylated PLN; WB, Western blot.

on Decem

ber 28, 2020http://stke.sciencem

ag.org/D

ownloaded from

Page 6: BIOCHEMISTRY Copyright © 2020 14-3-3 binding creates a ... · 9/1/2020  · Center Göttingen, Robert-Koch-Straße 40, 37075 Göttingen, Germany 7Max-Planck Institute for Biophysical

Menzel et al., Sci. Signal. 13, eaaz1436 (2020) 1 September 2020

S C I E N C E S I G N A L I N G | R E S E A R C H A R T I C L E

6 of 16

away from the phosphate group. In contrast, there is no arginine close to the phosphoresidue for pThr17-PLN in complex with 14-3-3 (Fig. 4B). Instead, Arg14 is pointing away from the phospho-residue consistent with some previously characterized 14-3-3/phosphopeptide complexes (36) and Arg13 contacts the solvent. Hence, its position is not reliable in the current model. Calculation of the peptide’s delta energy of binding to 14-3-3 along MD trajec-tories confirmed the binding assay data, showing that pThr17-PLN has a greater affinity for the protein than the pSer16-PLN peptide (−9.17 ± 2.17 and −4.43 ± 1.22 kcal/mol, respectively). Although the precise atomic determinants of the different stabilities await further characterization, modification of PLN by either PKA or CaMKII determines the conformation in which 14-3-3 binds to the resulting phosphoform and therefore the stability of the complex.

Disease-causing ∆Arg14 PLN recruits 14-3-3 exclusively through pThr17

Arg13 and Arg14 are multifunctional residues in the PLN N terminus because they are also essential elements of an Arg-based COPI- interacting motif (18). ∆Arg14 PLN leaves the endoplasmic reticulum,

whereas WT PLN is retained (37). In addition, Arg14 is crucial for the PKA consensus motif, resulting in the phos-phorylation of Ser16. However, Arg14 is lost as a consequence of a trinucleotide deletion in an inherited form of DCM (5). We concluded that this disease- associated variant of PLN lacks the low-affinity 14-3-3–binding site pro-vided by pSer16-PLN (Fig. 3C and fig. S6). However, we found that the CaMKII- dependent binding site provided by pThr17 was not altered in this disease- causing variant because we observed robust phosphorylation and 14-3-3 binding to pThr17-∆Arg14-PLN (Fig. 4D). A second disease variant associated with familial forms of DCM is R9C (7), which is not predicted to participate directly in the 14-3-3–binding phospho-peptide motif. Accordingly, mutating Arg9 to cysteine or alanine did not affect the binding of pThr17-PLN to 14-3-3 (Fig. 4D). These results imply that CaMKII-mediated phosphorylation of Thr17 is the only way by which PLN can recruit 14-3-3 in ∆Arg14-carrying patients.

14-3-3 slows the kinetics of PLN dephosphorylation14-3-3 binding has been suggested to protect against dephosphorylation by steric masking (23, 38). We measured the effect of 14-3-3 binding on the kinetics of the dephosphorylation of immobilized pThr17-PLN-GST or pSer16- T17A-PLN-GST by alkaline phosphatase (Fig. 5A and fig. S7A). We observed a substantial increase in the lifetime of

pThr17 or pSer16 when 14-3-3 was present (Fig. 5B and fig. S7B). Concomitantly with dephosphorylation, binding of 14-3-3 slowly decreased from either PLN phosphoform as expected (Fig. 5, C and D, and fig. S7A). We conclude that occupancy by 14-3-3 controls phosphatase access and thereby the lifetime of the posttranslational modification on either Ser16 or Thr17 of PLN.

Signal transduction recruits 14-3-3 to PLN-containing membranes in cardiomyocytesPLN is an abundant cardiac membrane protein and a major target of PKA-mediated regulation during -adrenergic stimulation (25). To determine the cellular consequences of PLN phosphorylation and 14-3-3 binding, we stimulated -adrenergic signal transduction in isolated adult rat cardiomyocytes using the agonist isoprenaline and assessed whether the predicted increase in pSer16-PLN led to the recruitment of 14-3-3. The N terminus of PLN is short, and 14-3-3 may occlude the antibody epitope. Nevertheless, we generated a rabbit antibody that immunoprecipitated SDS-resistant PLN pen-tamers (Fig. 6A). Detection of precipitated PLN used commercially available antibodies against PLN raised in mouse and against 14-3-3

A

PLNpPLN

PLN

kDa

Phostag-PAGE

3426

WB: 14-3-3 pan H8

CaMKII

3426

SDS-PAGE

3426

− +R9C

− +W

T

− +

C D

PLN pThrB

Coo

mas

sie

− +R9A

14-3-3

Pro

14

21

Met20 Glu19 Thr17

Arg14

Ala15Ser16

Ile18Pro21 Met20Ile18

Ala15Arg13

Arg14

Ser16

Thr17Glu19

17PLN pSer16

Fig. 4. 14-3-3 contacts PLN in two distinct conformations. (A) Overview of a 14-3-3 monomer bound to pSer16-PLN based on the model obtained by MD simulations. 14-3-3 is shown as a colored surface according to the electrostatic potential computed by APBS (negative potential in red, positive in blue), and peptides are depicted as cyan sticks (n = 2 independent experiments with conformational sampling of 250 ns). (B) Overview of a 14-3-3 monomer bound to pThr17-PLN as obtained by MD simulations. Visualization as in (A) (n = 2 independent experiments with conformational sampling of 250 ns). (C) Superimposition of 14-3-3/pSer16-PLN complex (white cartoon/cyan sticks) and 14-3-3/pThr17-PLN (pink cartoon/purple sticks) shown in (A) and (B). (D) In vitro phosphorylation of the indicated PLN-GST variants mimicking DCM-causing mutants using CaMKII (n = 5 independent experiments). pPLN, phosphorylated PLN; WB, Western blot.

on Decem

ber 28, 2020http://stke.sciencem

ag.org/D

ownloaded from

Page 7: BIOCHEMISTRY Copyright © 2020 14-3-3 binding creates a ... · 9/1/2020  · Center Göttingen, Robert-Koch-Straße 40, 37075 Göttingen, Germany 7Max-Planck Institute for Biophysical

Menzel et al., Sci. Signal. 13, eaaz1436 (2020) 1 September 2020

S C I E N C E S I G N A L I N G | R E S E A R C H A R T I C L E

7 of 16

raised in rabbit. In HeLa cell and total ventricular lysate, this anti-body detected two bands most likely reflecting 14-3-3 [its primary sequence is ~10 residues longer than that of all other isoforms (39)] (Fig. 6B). Untreated and isoprenaline-treated ventricular cardio-myocytes (Fig. 6C, panels labeled “Total”) showed comparable levels of PLN in the membrane fraction (Fig. 6D). Membranes from isoprenaline-treated cardiomyocytes showed a substantial increase in pSer16-PLN (Fig. 6E) and modestly increased levels of 14-3-3 (Fig. 6F). We performed coimmunprecipitation experiments, which suggested that endogenous PLN was associated with endogenous 14-3-3 proteins (Fig. 6C, panels labeled “IP: PLN”). Whether this association is direct cannot be definitively concluded from these experiments, although this interpretation is supported by the direct binding shown in vitro (Fig. 3, C and D) and by the preferred bind-ing of 14-3-3 dimers to SDS-resistant PLN pentamers, suggesting an avidity effect (Fig. 3, A and B).

We sought to further corroborate the idea that isoprenaline stimulation resulting in an increase in pSer16-PLN might elicit

biological effects by 14-3-3 binding and ensuing stabilization of the phosphoform of PLN as observed in vitro (Fig. 5A). Adult cardio-myocytes were patch-clamped, and Ca2+ transients were evoked at 0.5 Hz (fig. S8A). As expected, acute (10 s) stimulation with the adrenergic agonist isoprenaline elicited an apparent accelerated recovery of Ca2+ transients as quantified by the time constant tau of Ca2+ transient decay (fig. S8B) due to PKA-dependent PLN phos-phorylation and SERCA stimulation. We dialyzed cells with an intracellular pipette solution containing recombinant 14-3-3, which slowed the recovery of the time constant decay of the Ca2+ tran-sients back to baseline values as observed before isoprenaline stimulation (Fig. 7A and fig. S8, C and D). This effect was blocked by R18 (Fig. 7B), suggesting that it depended on the interaction of 14-3-3 with a target protein through its ligand-binding groove. To support the notion that this effect of 14-3-3 occurred through PLN and SERCA, we showed that, in mouse myocytes at 35°C, the con-tribution of the Na+/Ca2+ exchanger was negligible and that SERCA dominated the Ca2+ transient recovery (fig. S8, E to G).

FastAP(min)

25

CaMKII

PLNcyt-linker-GST WT

14-3-3 +

++ +− + −

+ +− +

+ +− +kDa

PLN

PLNp-PLN

1050

3426

3426

3426

34

26

A B

C

SDS-PAGE

Phostag-PAGE

Coo

mas

sie

CaMKII

PLNcyt-linker-GST WT

14-3-3 + +

++ +− + −

+ +− +

+ +− +kDa

251050

14-3-3 H8

D

FastAP(min)

0 5 10min FastAP treatment

25

Rel

ativ

e P

LN

pThr

in

tens

ity17

pThr17

+

Nor

m 1

4-3-

3 in

tens

ity0

0.20.40.60.81.01.21.4

00.20.40.60.81.01.2

−14-3-3 bound to PLN

0FastAP treatment (min)

5 10 25

+14-3-3 bound to PLN

Fig. 5. 14-3-3 protects PLN from dephosphorylation. (A) Dephosphorylation of pThr17-PLN-GST in the presence of 14-3-3 as monitored by anti–pThr17-PLN Western blot (n = 6 independent experiments). (B) Quantification of experiments shown in (A) using values of the PLN pThr17 intensity at different time points normalized to the PLN pThr17 intensity at time point 0 (no 14-3-3). (C) Experiments as shown in (A) illustrating the phosphorylation state of PLN-GST (phostag gel), total PLN-GST (SDS-PAGE), and bound 14-3-3 (Western blot). (D) Quantification of bound 14-3-3 from experiments shown in (A) and (C) using values of the 14-3-3 intensity at different points normalized to the 14-3-3 intensity at time point 0. Data in (B) and (D) are presented as box and whisker plots with mean and median. A paired t test showed no significant difference between the samples with or without 14-3-3 at 0 min (P = 0.2141). A significant difference was shown between the samples with or without 14-3-3 at 5 min (*P = 0.0181), at 10 min (*P = 0.0136), and at 25 min (*P = 0.0370). FastAP, alkaline phosphatase; min, minutes; Norm, normalized.

on Decem

ber 28, 2020http://stke.sciencem

ag.org/D

ownloaded from

Page 8: BIOCHEMISTRY Copyright © 2020 14-3-3 binding creates a ... · 9/1/2020  · Center Göttingen, Robert-Koch-Straße 40, 37075 Göttingen, Germany 7Max-Planck Institute for Biophysical

Menzel et al., Sci. Signal. 13, eaaz1436 (2020) 1 September 2020

S C I E N C E S I G N A L I N G | R E S E A R C H A R T I C L E

8 of 16

To test whether the effect of 14-3-3 on the recovery of the time constant could be explained by binding to pSer16-PLN, we com-pared the effect of two phospho-specific antibodies directed against pSer16-PLN or pThr17-PLN (Fig. 7C and fig. S1, C and D). Although the latter did not affect recovery, the pSer16 antibody led to a pro-longed slowing of the recovery of the time constant, similar to that caused by 14-3-3. During isoprenaline stimulation, pSer16 PLN is expected to be the dominant phosphoform of PLN. Furthermore, pThr17-PLN may already be saturated with endogenous 14-3-3 because of the higher binding affinity (Fig. 3, E to H). We predict that the pThr17 antibody or 14-3-3 would prolong the lifetime of pThr17-PLN and hence SERCA stimulation under conditions in which PLN becomes acutely and abundantly phosphorylated at this site. The comparison of the effect of the two antibodies during -adrenergic stimulation and concomitant PKA activation suggests

that pSer16-dependent, acute recruitment of a large antibody protein or 14-3-3 to PLN can affect the biological outcome of -adrenergic stimulation by prolonging the isoprenaline-stimulated state of SERCA in cardiomyocytes. Together, we predict that affinity and avidity effects combine to fine-tune the stability of PLN-14-3-3 complexes (Fig. 7D), resulting in different degrees of protection from phos-phatase action.

DISCUSSIONPLN is a key regulator of SERCA. Inherited disorders that result in PLN dysfunction demonstrate a critical role of this regulation in humans. Despite its short sequence, the N terminus of PLN is emerging as a hub where interactions with proteins distinct from the kinases and phosphatases that modify PLN in response to cardiac

14-3-3(14122)

180130

95

55

43

3426

72

kDa Heart

HeLa

IP: PLN (r-IgG)

3426

17

PLN (m-IgG)

kDa Inpu

t 10%

Ig

G

PLN

A

B

C D

E

F

M

P

*

− IS

O+

ISO

0

0.5

1.0

1.5

14-3

-3/N

aK *

− IS

O+

ISO

− IS

O+

ISO

0

0.5

1.0

1.5

PLN

/ NaK

P

LN p

Ser

/

PLN

16

Pln−/−

WB: Na/K ATPase95

130

kDa

WB: 14-3-3 14122

2634

WB: PLN (m-IgG)

2634

WB: 14-3-3 14122

WB:PLN pSer16

2634

WB: PLN (m-IgG)

2634

2634

Pln

− + ISO1

− + − + − +IP: PLN(r-IgG)

Total

IP: PLN(r-IgG)

*

Total

Total

Total

2 3 4

−/− Pln+/+

WB: Na/K ATPase95

130 Total

0

1

2

3 *

Fig. 6. 14-3-3 is enriched in mouse cardiac membranes after isoprenaline stimulation. (A) Immunoprecipitation from cardiac membranes using PLN (rabbit IgG) antibody. Detection with PLN (mouse IgG) antibody (n = 2 mice per group). (B) 14-3-3 pan-14122 antibody characterization using HeLa cell and heart lysate. Upper band indicates a slower migrating 14-3-3–positive band, presumably 14-3-3 epsilon based on predicted molecular weights. Asterisk indicates cross-reacting protein (n = 2 mice or independently transfected cell populations per group). (C) Detection of PLN-associated 14-3-3 after isoprenaline stimulation in cardiomyocytes. IP, immunoprecipitation of PLN with PLN (rabbit IgG) antibody from isoprenaline-stimulated cardiomyocytes; Total, 15% input membranes. Top two panels: anti-14-3-3 detection of PLN IP (run without DTT) and cardiac inputs (run with DTT); middle three panels: anti-PLN (mouse IgG) or anti-Na/K detection of PLN IP (run without DTT) or cardiac inputs (run with DTT); lower two panels: anti–pSer16-PLN or anti-Na/K detection of cardiac inputs (n = 4 mice per group). Asterisk indicates cross-reacting protein. (D to F) Quantification of experiments shown in (C). Data are presented as box and whisker plots with mean and median. A paired t test showed no significant difference in (D) (P = 0.8771), but a significant difference was found in (E) (*P = 0.0338) and in (F) (*P = 0.0222). 14-3-3 detected in membranes from Pln−/− mice (negative control for IP) is shown in purple in (F). IP, immunoprecipitation; IgG, immunoglobulin G; m-IgG, mouse immunoglobulin; r-IgG, rabbit immunoglobulin; M, monomer; P, pentamer; WB, Western blot; total, total amount of protein used as input; NaK, Na/K ATPase; ISO, isoprenaline.

on Decem

ber 28, 2020http://stke.sciencem

ag.org/D

ownloaded from

Page 9: BIOCHEMISTRY Copyright © 2020 14-3-3 binding creates a ... · 9/1/2020  · Center Göttingen, Robert-Koch-Straße 40, 37075 Göttingen, Germany 7Max-Planck Institute for Biophysical

Menzel et al., Sci. Signal. 13, eaaz1436 (2020) 1 September 2020

S C I E N C E S I G N A L I N G | R E S E A R C H A R T I C L E

9 of 16

signal transduction may enhance or attenuate the ability of PLN to inhibit SERCA (16). We identified 14-3-3 phosphoadaptor proteins as PLN interactors through several independent lines of evidence.

First, 14-3-3 affinity purification enriched SDS-resistant PLN pentamers from solubilized cardiac membranes (Figs. 1E and 3A). APEX2 peroxidase–based proximity labeling demonstrated that

C

−200 −100 0 100 200 300 400 500

−1.25

−1.00

−0.75

−0.50

−0.25

0.00

0.25

Time (s)

Control

−200−100 0 100 200 300 400 500

−1.25

−1.00

−0.75

−0.50

−0.25

0.00

0.25

Time (s)

pThr Control

pSer ISO

A B

C D

pSer pThr

C

R18Control

ISO ISO

−200 −100 0 100 200 300 400 500

−1.25−1.00−0.75−0.50−0.25

0.000.25

Time (s)

17

16AbAb

16 17

*

*

C

Affinity Avidity

Accessible Protected

C

C

pSer1614-3-3PP1 PP1

Fig. 7. 14-3-3 prolongs SERCA stimulation after -adrenergic activation in mouse cardiomyocytes. The data in (A) to (D) were obtained under control conditions or in the presence of the tested peptides or antibodies in the patch pipette solution. Ca2+ transients were recorded at a 0.5-Hz stimulation rate, as shown in fig. S8 (A and C). The time constant () of Ca2+ transient decay was measured every 55 s. To obtain an individual reading, we averaged values calculated by single exponential fitting of decay phase for five Ca2+ transients recorded at each time point. The changes in the SERCA function followed the application of 10 nM isoprenaline for 10 s. To compen-sate for the cell-to-cell variability in the absolute magnitude of isoprenaline peak response, the relative response from baseline was normalized to the peak response in each cell. (A) Ca2+ transient decay in cardiomyocytes patched with control pipette solution (seven cells from four isolations) or with pipette solution containing 14-3-3 (2 M) (n = 7 cells from five isolations). A two-way ANOVA mixed-model analysis revealed significant differences between curves (P = 0.0024). A multiple comparisons test showed significant differences only at the last time point (*P = 0.033). (B) Ca2+ transient decay in cardiomyocytes patched with control pipette solution (seven cells from four isolations) or with pipette solution containing R18 (2 M) (n = 8 cells from four isolations) or a mixture of 14-3-3 (2 M) and R18 (2 M) (n = 6 cells from four isolations). Two-way ANOVA indicated no significant differences between the three tested conditions (control, R18 inhibitor, and 14-3-3 blocked with R18) either using mixed-model analysis (P = 0.6365). (C) Ca2+ transient decay in cardiomyocytes patched with control pipette solution (n = 7 cells from four isolations) or with pipette solution containing 2.5 g/ml of either anti-pSer16 (n = 6 cells from five isolations) or anti-pThr17 antibody (n = 6 cells from four isolations). A two-way ANOVA mixed-model analysis revealed significant differences between the curves of the three tested conditions (P = 0.0131). Multiple comparisons tests showed significant differences between anti-pSer16 and anti-pThr17 antibodies at 380 s (*P = 0.0143) and further 440 s (*P = 0.0053). Between control and anti-Ser16 group, significant differences were observed at 380 s (*P = 0.0495) and at 440 s (*P = 0.0268). All other changes were not significant. (D) Scheme illustrating the dependence of the apparent binding affinity of dimeric 14-3-3 protein on the number and kind of different phospho-residues present in the PLN pentamer. Affinity and avidity effects combine with each other, resulting in increased 14-3-3 binding to PLN and protection from the phosphatase PP1. ISO, isoprenaline; s, seconds; R18, R18 trifluoroacetate; Ab, antibody; PP1, protein phosphatase 1.

on Decem

ber 28, 2020http://stke.sciencem

ag.org/D

ownloaded from

Page 10: BIOCHEMISTRY Copyright © 2020 14-3-3 binding creates a ... · 9/1/2020  · Center Göttingen, Robert-Koch-Straße 40, 37075 Göttingen, Germany 7Max-Planck Institute for Biophysical

Menzel et al., Sci. Signal. 13, eaaz1436 (2020) 1 September 2020

S C I E N C E S I G N A L I N G | R E S E A R C H A R T I C L E

10 of 16

14-3-3 proteins are localized within nanometers of PLN in living cardiomyocytes (Fig. 2B). We coimmunoprecipitated PLN with 14-3-3 (Fig. 6C) from isolated cardiomyocytes. Although these complementary approaches do not demonstrate a direct interaction between PLN and 14-3-3, we argue that a direct interaction is plausible because Ser16 and Thr17 are located in putative 14-3-3 binding sites (18, 34), because we showed a direct interaction in a phosphorylation- dependent manner (Fig. 3, C and G), and because MD simulations yielded plausible structural models for the respective PLN-derived phosphopeptide in the ligand-binding groove of 14-3-3 (Fig. 4, A to C).

14-3-3 proteins are dimeric and, hence, susceptible to avidity effects when binding to multimeric targets. We found that SDS-resistant pentameric, but not monomeric, PLN bound 14-3-3 when phos-phorylated by PKA (Figs. 1E and 3A). The SDS-resistant PLN pentamers were previously interpreted to be proxies of physiological pentamers (9, 14). Even if our binding experiment performed in a mild detergent (ComplexioLyte 48) and subsequent analysis by SDS-PAGE captured only a fraction of all PLN pentamers, the experiment showed that PLN pentamers could recruit 14-3-3. However, we reconstituted the interaction between CaMKII- phosphorylated PLN and 14-3-3 using a recombinant non-oligomeric protein (Fig. 3, G and H). Hence, we propose a model (Fig. 7D) in which the relative abundance of pSer16 and pThr17 phosphosites determines the occupancy of PLN pentamers by 14-3-3. Acute PKA-mediated phosphorylation of PLN during -adrenergic stim-ulation is expected to raise the apparent affinity for 14-3-3. We observed prolonged SERCA activity due to 14-3-3 binding after -adrenergic stimulation in cardiomyocytes (Fig. 7A). Because 14-3-3 binding slowed the dephosphorylation of PLN phosphosites in vitro (Fig. 5, A and B, and fig. S7, A and B), we propose that the biological effects of 14-3-3 binding are due to blocking the access of PP1c to the phosphosite, thereby maintaining PLN in the inactive, phosphorylated state and SERCA in the active state. Consistently, we mimicked the effect of 14-3-3 after acute -adrenergic stimula-tion in cardiomyocytes by the addition of a specific anti–pSer16-PLN antibody (Fig. 7C). Binding of 14-3-3 dimers to the pSer16-PLN pentamers may also stabilize the pentameric form of PLN and hence reduce the pool of monomeric form that is thought to be the major species that inhibits SERCA (15).

The affinity of the respective protein-protein interaction between 14-3-3 and pSer16- or pThr17-PLN translates into a molecular “memory” of which kinase created the 14-3-3–binding site on PLN. At the molecular and atomic level, this memory is because the respective phosphoresidues, which are specific products of either PKA or CaMKII, fit into the same position of the 14-3-3 ligand-binding groove (Fig. 4, A and B). Consequently, the sur-rounding side chains of the PLN N terminus slot into different, energetically more or less favorable positions of the binding groove. For example, comparison of pSer16-PLN with pSer16-T17A-PLN (Fig. 3, C and D) indicated that the side chain of Thr17 reduced the energy of binding when Ser16 was phosphorylated. When Thr17 was phosphorylated, Ile18 occupied the corresponding position, which avoided placing the polar side chain of Thr17 in this environment and hence resulted in an energetically more favorable fit.

Biologically, a shorter memory for PKA-mediated phosphoryla-tion is consistent with the role of this modification in the acute and reversible adaptation of intracellular Ca2+ handling to the cAMP (adenosine 3′,5′-monophosphate) signal elicited by -adrenergic agonists. CaMKII activation in cardiomyocytes integrates changes

in intracellular Ca2+ concentrations and other parameters such as redox potential (40). CaMKII activity can become independent of acute changes in intracellular Ca2+. The kinase has many targets in addition to intracellular Ca2+ handling proteins, such as ion channels that control cardiomyocyte excitation and proteins that mediate the transcriptional activation of a hypertrophic gene program. The slower effects of CaMKII on gene expression could represent a biological rationale for the longer-lasting memory associated with PLN modification by CaMKII kinase, which translates to a higher affinity of 14-3-3 binding to pThr17-PLN. This mechanism may ensure the adaptation of intracellular Ca2+ handling to the hyper-trophic state. Expression of CaMKII itself is elevated in heart failure (41), and the kinase has attracted attention as a potential therapeutic target (42). Our insight into CaMKII-dependent 14-3-3 recruitment to PLN may help to understand the effects of CaMKII inhibitors that are currently under development.

Two forms of genetic PLN-dependent DCM alter the N terminus of PLN with different effects on phosphorylation and 14-3-3 binding. Our data can therefore help to extend current models (7, 43, 44) of the molecular basis of ∆Arg14- and R9C-PLN. On the one hand, ∆Arg14-PLN cannot be modified by PKA (fig. S6) and hence lacks the low-affinity 14-3-3–binding mode. This may make CaMKII activation upon sustained -adrenergic stimulation and 14-3-3 binding to PLN through pThr17 more crucial. On the other hand, the introduction of a reactive cysteine side chain instead of Arg9 leads to disulfide bridges that stabilize the PLN pentamer and render it less accessible to PKA (44). Access of 14-3-3 to the pertinent phosphopeptide in such disulfide-stabilized PLN pentamers would also be impaired, resulting in an inability to differentially stabilize the pentamer according to kinase action. The precise consequences of either DCM mutation on 14-3-3 recruitment to PLN will have to be biochemically investigated in the corresponding heterozygous disease models.

In conclusion, PLN represents yet another fascinating example in which the action of 14-3-3 proteins modulates the outcome of signal transduction events by recognizing the phosphorylation and oligomerization state of a target protein. Steric masking of the phos-phosite and potential stabilization of the pentamer through 14-3-3 binding may influence not only the access of other proteins such as phosphatases but also the dynamics of PLN itself. In line with the different delta energies of binding in the two molecular registers of 14-3-3 binding, these effects and their duration seem to be fine-tuned according to the kinase that modified PLN.

MATERIALS AND METHODSPurification of proteinsRecombinant GST- or myelin basic protein (MBP)–tagged proteins (expressed from the plasmids listed in table S1) were expressed and purified from Escherichia coli strain BL21 Rosetta. Protein expres-sion was induced in 2YT medium (2YT mix, AppliChem) using 1 mM isopropyl--d-thiogalactopyranoside (IPTG) for 3 hours at 30°C. Cells were harvested and lysed in GST buffer [20 mM Hepes (pH 6.8), 150 mM KOAc, 5 mM Mg(OAc)2, 1 mM EDTA, 1 mM dithiothreitol (DTT), and 1 mM phenylmethylsulfonyl fluoride (PMSF; pH 6.8)]. Crude cell lysate was centrifuged at 100,000g for 30 min at 4°C. The supernatant was incubated with 1 ml of washed glutathione Sepharose beads (GE Healthcare) for GST-tagged proteins or 1 ml of amylose resin (New England Biolabs) for MBP-tagged

on Decem

ber 28, 2020http://stke.sciencem

ag.org/D

ownloaded from

Page 11: BIOCHEMISTRY Copyright © 2020 14-3-3 binding creates a ... · 9/1/2020  · Center Göttingen, Robert-Koch-Straße 40, 37075 Göttingen, Germany 7Max-Planck Institute for Biophysical

Menzel et al., Sci. Signal. 13, eaaz1436 (2020) 1 September 2020

S C I E N C E S I G N A L I N G | R E S E A R C H A R T I C L E

11 of 16

proteins for 1 hour at 4°C under gentle rotation. The bead slurry was transferred to gravity columns and washed with three column volumes of GST buffer (pH 7.4), one column volume of 1 mM ATP in GST buffer (pH 7.4), and GST buffer (pH 7.4). Bound GST-tagged proteins were eluted with GST elution buffer [20 mM Hepes (pH 6.8), 150 mM KOAc, 5 mM Mg(OAc)2, 1 mM EDTA, 1 mM DTT, and 15 mM l-glutathione (pH 8.5)]. Bound MBP-tagged proteins were eluted with MBP elution buffer [20 mM Hepes (pH 6.8), 150 mM KOAc, 5 mM Mg(OAc)2, 1 mM EDTA, 1 mM DTT, and 20 mM d-maltose (pH 7.5)]. The MBP tag was cleaved with FactorXa enzyme for 16 hours at 25°C according to the supplier’s recommendation (Merck Millipore). After cleavage, proteins were separated by size exclusion chromatography on an Äkta purifier (GE Healthcare) and a Superdex75 size exclusion column in GST buffer (pH 7.5).

Heart membrane preparationAfter defrosting on ice, fat tissue was removed from mouse hearts, which were then washed in tissue homogenization buffer [50 mM NaCl, 320 mM sucrose, 2 mM EDTA, and 20 mM Hepes (pH 7.4)] supplied with protease and phosphatase inhibitors (Roche). Hearts were cut into small pieces and lysed with a MiccraD-1 homog-enizer for 30 s followed by douncing for 30 strokes. Cytosol and membranes were separated by centrifugation at 100,000g for 30 min at 4°C. The supernatant was the cytosolic fraction. The purified membrane fraction was resuspended once, washed in homogenization buffer, and collected by centrifugation at 100,000g for 30 min.

Heart membrane solubilizationPurified heart membranes were solubilized in ComplexioLyte 48 (Logopharm) buffer for 30 min on ice. ComplexioLyte 48 (1 ml) was used to solubilize membranes with a protein content of 1 mg. Solubilized membranes were centrifuged for 30 min at 55,000g, 4°C. The supernatant contained solubilized proteins, which were used for pull-down experiments or trichloroacetic acid (TCA) precipitated for Western blot analysis.

Separation of PLN pentamer into monomer in cardiac membranes by heatingMouse heart membranes were purified as described above. Solubilized mouse heart membranes were diluted 1:5 in GST buffer [20 mM Hepes (pH 6.8), 150 mM KOAc, 5 mM Mg(OAc)2, 1 mM EDTA, 1 mM DTT, and 1 mM PMSF (pH 7.4)] and divided into two Eppendorf tubes. One tube was stored on ice, and the other one was heated at 75°C for 20 min. After heating, the membranes were centrifuged at 20,000g for 10 min at 4°C. Both solubilized mem-brane preparations were used as inputs for pull-down experiments (Fig. 3, A and B).

Phostag polyacrylamide gel electrophoresisPhostag acrylamide (NARD Institute) was used at a concentration of 5 M according to the supplier’s information.

Western blot detectionPrimary antibodies were diluted 1:1000 in blocking buffer [5% (w/v) milk powder, 1× TBS, and 0.1% Tween] and incubated overnight at 4°C or for 2 to 3 hours at room temperature. Western blots were imaged with an Odyssey CLx imaging system using IRDye

LI-COR secondary antibodies. Secondary antibodies were diluted 1:10,000  in blocking buffer and incubated for 1 hour at room temperature.

14-3-3 pull-down experiments from mouse cardiac membranes and cytosolSolubilized cardiac membranes or cytosol and recombinant puri-fied MBP or MBP-14-3-3 bait protein was used for pull-down experiments. MBP or MBP-14-3-3 protein (50 g) was immobilized to 5 l of amylose resin bead slurry (New England Biolabs) in GST buffer [20 mM Hepes (pH 6.8), 150 mM KOAc, 5 mM Mg(OAc)2, 1 mM EDTA, and 1 mM DTT (pH 7.5)] supplied with protease and phosphatase inhibitors (Roche) for 45 min at 4°C under gentle rotation. Immobilized bait protein was washed three times in GST buffer, followed by blocking in 4% bovine serum albumin dissolved in GST buffer, and supplied with protease and phosphatase inhibitors (Roche) for 30 min at 4°C under gentle rotation to block unspecific binding sites. For the MBP–14-3-3 protein, 5 M R18 trifluoro-acetate was supplied during immobilization and blocking. After blocking, the resin-bound bait was washed three times in GST buffer and added to prepared solubilized cardiac membranes or cytosol, which were previously diluted 1:5 in GST buffer (pH 7.5). After a 1-hour incubation at 4°C under gentle rotation, the resin- bound 14-3-3 was washed four times in GST buffer followed by elution with 1× SDS sample buffer (containing 100 mM DTT). The pull-down experiment was analyzed by SDS-PAGE or Phostag- PAGE followed by silver staining or Western blot detection using primary antibodies described in table S2.

APEX2 proximity labeling in NRCMsNRCMs were isolated from 0- to 3-day-old Wistar rats, and 5 × 105 cells per well were seeded on six-well plates. NRCMs were cultivated in heavy, medium, and light (Eurisotop) SILAC (stable isotope labeling by amino acids in cell culture) medium (Thermo Fisher Scientific) under humidified conditions at 37°C. After 13 days, NRCMs were transduced with adenoviral vectors [V5-APEX2–PLN, V5-APEX2–PLN(1–29), or eGFP (enhanced GFP)] with a multiplicity of infection of 10 for 48 hours. The proximity labeling was conducted as described elsewhere (45). Briefly, the cells were incubated for 30 min with 0.5 mM biotin-phenol at 37°C followed by 1 mM H2O2 for 1 min at room temperature. Cells were washed with quenching buffer [5 mM Trolox, 10 mM NaN3, and 10 mM Na-ascorbate in phosphate-buffered saline (PBS)] and harvested in radioimmunoprecipitation assay buffer [0.5% Na-deoxycholate, 50 mM tris-HCl (pH 7.5), 150 mM NaCl, 0.2% SDS, 1% Triton, 10 mM NaN3, 5 mM Trolox, 10 mM Na-ascorbate, 1 mM PMSF, and com-plete mini]. Cells were lysed by passing 10 times through a 27-gauge needle and centrifuged at 13,000g for 10 min. Protein concentra-tions were determined with the Pierce 660nm Protein Assay Kit (Thermo Fisher Scientific). The lysates of heavy-, medium-, and light-labeled NRCM were mixed in a ratio of 1:1:1 for a total of 250 g of protein. Biotinylated proteins were enriched by avidin pull-down assay (Thermo Fisher Scientific). Eluted proteins were analyzed by liquid chromatography–tandem mass spectrometry (LC-MS/MS) and MaxQuant. For quantification, log2 fold changes of ratios of PLN/eGFP and PLN/PLN(1–29) were calculated and proteins that differed positively from 0 according to a one-sample t test (P < 0.05) were plotted. The experiment was conducted in three biological and two technical replicates with label switch.

on Decem

ber 28, 2020http://stke.sciencem

ag.org/D

ownloaded from

Page 12: BIOCHEMISTRY Copyright © 2020 14-3-3 binding creates a ... · 9/1/2020  · Center Göttingen, Robert-Koch-Straße 40, 37075 Göttingen, Germany 7Max-Planck Institute for Biophysical

Menzel et al., Sci. Signal. 13, eaaz1436 (2020) 1 September 2020

S C I E N C E S I G N A L I N G | R E S E A R C H A R T I C L E

12 of 16

LC-MS/MS analysisProteins were separated on 4 to 12% bis-tris minigels, visualized with Coomassie blue stain, and sliced into 11 sections of equal size regardless of staining. After reduction with DTT, alkylation with 2-iodoacetamide, and overnight digestion with trypsin, digest pep-tide mixtures were extracted and dried in a speedvac (46). For MS analysis, samples were enriched on a self-packed reversed-phase C18 precolumn [0.15 mm inside diameter (ID) × 20 mm, Reprosil- Pur120 C18-AQ 5 m, Dr. Maisch] and separated on an analytical reversed-phase C18 column (0.075 mm ID × 200 mm, Reprosil-Pur 120 C18-AQ, 3 m, Dr. Maisch) using a 30-min linear gradient of 5 to 35% acetonitrile/0.1% formic acid (v:v) at 300 nl min−1. The eluent was analyzed on a Q Exactive hybrid quadrupole/orbitrap MS (Thermo Fisher Scientific) in data-dependent acquisition mode. Each experimental cycle contained a full MS scan [350 to 1600 mass/charge ratio (m/z), resolution setting 70,000 full width at half maximum (FWHM), automatic gain control (AGC) target 1 ×106, maximum fill time of 60 ms] and up to 12 MS/MS experiments (z = 2 to 5, 2 ×104 trigger threshold, dynamic exclusion window 15 s, 2.0 FWHM iso-lation width, normalized collision energy 25%, resolution setting 17,500 FWHM, AGC target 2 × 105, maximum fill time 60 ms). Two technical replicates per sample were acquired. Raw data were processed using MaxQuant software version 1.5.2.8 (Max Planck Institute for Biochemistry). The arginine R10/R6 and lysine K8/K4 labels including the “requantify” option were specified for relative protein quantitation.

ImmunofluorescenceNRCMs were isolated from 1- to 3-day-old Wistar rats and plated on coverslips coated with collagen. After 15 days of cultivation and 48 hours after transduction, cells were fixed in 4% paraformaldehyde for 5 min. Next, cells were incubated in blocking/permeabilization buffer for 4 hours (10% bovine calf serum and 0.2% Triton in PBS). Primary antibodies were diluted in blocking buffer and incubated overnight at 4°C as follows: V5 (1:1000, R960-25, Invitrogen) and SERCA2a (1:500, A010-20, Badrilla). Subsequently, samples were washed three times in blocking buffer and incubated with secondary antibodies diluted 1:1000 and incubated overnight at 4°C: anti- mouse (Alexa Fluor 568, Thermo Fisher Scientific) and anti-rabbit (Alexa Fluor 633, Thermo Fisher Scientific). After washing, samples were embedded in mounting medium (ProLong Gold antifade reagent, Thermo Fisher Scientific). For imaging, we used a Zeiss LSM 880 system with a Plan-Apochromat 63×/1.4 oil DIC objective. Raw images were processed in ImageJ/Fiji (http://imagej.nih.gov/).

Binding assay using purified PLN-GST constructs and 14-3-3Purified PLN-GST variants (10 g) were phosphorylated with re-combinant PKA catalytic subunit (New England Biolabs) or CaMKII (Thermo Fisher Scientific) according to the supplier’s information at 30°C for 1 hour. Phosphorylated bait proteins were bound to 5 l of glutathione Sepharose beads (GE Healthcare) in GST buffer [20 mM Hepes, 150 mM KOAc, 5 mM Mg(OAc)2, 1 mM EDTA, 1 mM DTT, and 0.1% Triton X-100 (pH 7.5)] for 45 min at 4°C under gentle rotation. Resin-bound bait proteins were incubated with equimolar amounts of recombinant purified 14-3-3 protein in GST buffer for 1 hour at 4°C under gentle rotation. The resin- bound bait was washed four times in GST buffer and eluted in 1× SDS sample buffer (supplied with 100 mM DTT). The binding assay

was analyzed on SDS-PAGE, which was transferred to nitrocellulose and probed with the indicated primary antibodies (compare table S2). Phosphorylation of bait proteins was analyzed on Coomassie-stained Phostag-PAGE and SDS-PAGE gels.

Surface plasmon resonanceBinding affinities of recombinant 14-3-3 protein to the cytosolic domain of recombinant purified PLN-GST were determined with surface SPR. SPR experiments were carried out on a Reichert SR 7500 DC biosensor instrument at 20°C and a flow rate of 40 l/min in SPR buffer [150 nM NaCl, 20 mM Hepes, and 0.005% (v/v) Tween 20 (pH 7.5)]. All experiments were done in a buffer containing 20 mM Hepes (pH 7.5), 150 mM NaCl, and 0.005% Tween 20 (v/v). An HC-1000 SPR sensor chip was activated with 1-ethyl-3-83- dimethylaminopropyl)-carbodiimide and N-hydroxysuccinimide (EDC-NHS) (Xantec Biotechnologies) according to the supplier’s instruction, followed by covalent binding of a GST antibody (Carl Roth, 3998.1) to the chip surface at a flow rate of 15 l/min and a concentration of 30 g/ml to a surface density of 4000 to 8000 RIU (refractive index units; depending on the maximal capacity of each chip). After antibody coupling, the chip surface was deactivated by quenching with 1 M ethanolamine (pH 8.5) for 180 s at a flow rate of 30 l/min. PLN-GST fusion proteins were phosphorylated with recombinant PKA catalytic subunit (New England Biolabs) or CaMKII (Thermo Fisher Scientific) kinase according to the manu-facturer’s instructions for 1 hour at 30°C and immobilized to the chip surface at a flow rate of 30 l/min and to a surface density of 200 to 400 RIU. The 14-3-3 analyte was serially diluted and injected over the sensor chip surface at a flow rate of 40 l/min. Association was done for 4.5 min, and dissociation was subsequently done for 7 min. The dilution series ranged from 0.390 mM to a final concentration of 50 M. Data analysis was performed with Scrubber 2.0c. Equilibrium binding isotherms were analyzed with GraphPad Prism 6.0.

Cardiomyocyte isolation and stimulation with isoprenalineAll animal procedures were reviewed and approved by the Institu-tional Animal Care and Use Committees of the University Medical Center Göttingen in compliance with the humane care and use of laboratory animals. Hearts were perfused by Langendorff solution [120.4 mM NaCl, 17.7 mM KCl, 0.6 mM KH2Po4, 0.6 mM Na2HPO4, 1.2 mM MgSO4, 10 mM Hepes, 4.6 mM NaHCO3, 30 mM taurine, 10 mM 2,3-butanedione-monoxime, and 5.5 mM glucose (pH 7.4)] with a flow of 4 ml/min for a period of 4 min at 37°C. To digest the tissue, the heart was perfused with perfusion buffer supplemented with collagenase type II (2 mg/ml) and CaCl2 (40 M) for 8 min. The heart was removed from the Langendorff perfusion apparatus and placed in collagenase solution, and the tissue was mechanically disrupted with a scissor and by resuspension with a pipette. The collagenase was stopped with bovine calf serum (10%) and CaCl2 (12.5 M). Isolated cardiomyocytes were collected by filtration with a 100 M cell sieve and centrifugation at 22g for 1.5 min. Isolated cardiomyocytes from one animal were split into two Eppendorf tubes in experimental buffer [10 mM Hepes, 136 mM NaCl, 40 mM KCl, 1 mM MgCl2, 10 mM glucose, and 1 mM CaCl2 (pH 7.35)], and one tube was treated with 100 nM isoprenaline for 30 s, whereas the other was not treated. After isoprenaline stimulation, the cells were collected by centrifugation and resuspend in sucrose homog-enization buffer [50 mM NaCl, 320 mM sucrose, 2 mM EDTA, and

on Decem

ber 28, 2020http://stke.sciencem

ag.org/D

ownloaded from

Page 13: BIOCHEMISTRY Copyright © 2020 14-3-3 binding creates a ... · 9/1/2020  · Center Göttingen, Robert-Koch-Straße 40, 37075 Göttingen, Germany 7Max-Planck Institute for Biophysical

Menzel et al., Sci. Signal. 13, eaaz1436 (2020) 1 September 2020

S C I E N C E S I G N A L I N G | R E S E A R C H A R T I C L E

13 of 16

20 mM Hepes (pH 7.4)] supplied with protease and phosphatase inhibitors (Roche) and stored at −80°C.

ImmunoprecipitationIsoprenaline (100 nM, 30 s)–stimulated cardiomyocytes were thawed, and membranes were separated as described above. Cardio-myocytes were lysed, and crude membranes were separated from the cytosolic fraction by centrifugation. Purified membranes were washed and solubilized in 500 l of ComplexioLyte 48 (Logopharm) for 30 min on ice. After solubilization, the samples were centrifuged at 55,000g for 30 min at 4°C. The solubilized membrane fraction (30%) was removed and complemented with 5× SDS sample buffer [250 mM tris-HCl (pH 6.8), 10% SDS, 0.5% bromophenol blue, and 50% glycerol] to a final concentration of 1× and analyzed by Western blotting detecting the indicated antigens using primary antibodies listed in table S2. The remaining solubilized membranes were diluted 1:5 in GST buffer [20 mM Hepes (pH 6.8), 150 mM KOAc, 5 mM Mg(OAc)2, 1 mM EDTA, and 1 mM DTT (pH 7.5)], and 5 l of PLN total antibody [rabbit immunoglobulin G (IgG)] was added. The antibody was incubated for 4 hours at 4°C under gentle rotation. Washed Sepharose A (GE Healthcare) was added to the samples and incubated overnight at 4°C under gentle rotation. Last, the beads were washed two times in GST buffer and eluted in 1× SDS sample buffer. The SDS sample buffer was not supplemented with DTT to avoid reducing conditions. The samples were separated on SDS-PAGE gels, transferred to nitrocellulose, and analyzed using the indicated primary antibodies.

Dephosphorylation of pThr17- or pSer16-T17A-PLN-GST with or without bound 14-3-3Recombinant purified PLN-GST protein (10 g) was used per sample and time point. Bait protein was phosphorylated with CaMKII (Thermo Fisher Scientific) or PKA (New England Biolabs) according to the supplier’s information at 30°C for 1 hour. Phosphorylated recombinant protein was bound to 5 l of glutathione Sepharose beads (GE Healthcare) in GST buffer [20 mM Hepes (pH 6.8), 150 mM KOAc, 5 mM Mg(OAc)2, 1 mM EDTA, 1 mM DTT, and 0.1% Triton X-100 (pH 7.5)] for 45 min at 4°C under gentle rotation followed by three wash steps in GST buffer. The immobilized bait protein was divided in two Eppendorf tubes. To one tube, equimolar amounts of recombinant purified 14-3-3 protein were added in GST buffer and allowed to bind for 1 hour at 4°C under gentle rotation. After binding, the bait proteins were washed four times in GST buffer. The resin-bound proteins were transferred to alkaline phosphatase buffer (Thermo Fisher Scientific) and treated with 8 U of alkaline phosphatase (Fast AP, Thermo Fisher Scientific) according to the supplier’s information for 0, 5, 10, and 25 min at 30°C. After each incubation time point, the resin-bound bait protein was washed two times in GST buffer and eluted in 1× SDS sample buffer (supplied with 100 mM DTT). The experiment was analyzed on SDS-PAGE or Phostag-PAGE. SDS-PAGE was transferred to nitro-cellulose membranes and analyzed for pSer16 or pThr17-PLN as appropriate using the antibodies listed in table S2. Phostag gels were stained with Coomassie.

Mouse cardiomyocyte isolation and cell culture for patch-clamp experimentsCells were isolated from hearts of 9- to 11-week-old male mice (c57bl6) by standard collagenase digestion as previously described

(47). All animal procedures were reviewed and approved by the Institutional Animal Care and Use Committees of King’s College London in compliance with the humane care and use of laboratory animals. Myocytes were then pelleted by gravity and washed at room temperature with modified M199 (Earle’s salts) culture medium (Thermo Fisher Scientific) containing (per 500-ml bottle) 0.113 g of creatine, 0.313 g of taurine, 0.198 g of carnitine, and 5 ml of penicillin/streptomycin (final concentration, 1%). Cells were pelleted, re-suspended in fresh medium, and transferred to laminin-covered coverslips (15-mm-diameter No1) in 12-well plates at an approximate final density of 150 myocytes per coverslip. Cells were incubated for 24 hours (37°C, 5% CO2) before use in electrophysiological experiments.

Measurement of intracellular [Ca2+] and patch-clamp experimentsMembrane currents were measured at 35°C in whole-cell ruptured- patch configuration using voltage-clamp with simultaneous intra-cellular Ca2+ measurement as previously described (48). Patch-clamp experiments were performed using an Axopatch 200B amplifier, Digidata 1322A data acquisition, and pClamp9.2 software (Molecular Devices, Sunnyvale, CA). Borosilicate glass microelectrodes had tip resistances of 1 to 2 megohms when filled with pipette solution containing EGTA (0.024 mM), Fluo-3 (0.025 mM), guanosine triphosphate (GTP)–tris (0.1 mM), Hepes (10 mM), K-aspartate (92 mM), KCl (48 mM), Mg-ATP (1 mM), and Na2-ATP (4 mM) (pH 7.25 at 35°C). During experiments, myocytes were superfused at 35°C with a Tyrode’s solution containing CaCl2 (1 mM), glucose (10 mM), Hepes (10 mM), KCl (4 mM), MgCl2 (1 mM), NaCl (136 mM), and probenecid (2 mM) (pH 7.35 at 35°C). Cells were kept at a holding potential of −80 mV, and Ca2+ transients were stimulated through the pipette at 0.5 Hz as previously described (2). In addition to including Fluo-3 in the pipette solution, myocytes were also preincubated with Fluo-3–acetoxymethyl ester [Invitrogen, Thermo Fisher Scientific; 1 mM solution in dimethyl sulfoxide (DMSO)]. Briefly, the cardiomyocyte staining solution was pre-pared by adding 5 l of Fluo-3 solution (1 mM; in DMSO) to 1 ml of bath solution and 160 l of deionized water to limit the osmotic effects of DMSO. Cardiomyocytes were incubated in staining solution for 40 min, followed by washing in bath solution (20 min) before use in voltage-clamp experiments. Optical measurements were performed using an Evolve 512 camera (Photometrix) and PTI EasyRatio Pro software (HORIBA Scientific). To avoid overloading the computer memory (due to high-speed acquisition at ~200 Hz), Ca2+ transients were recorded in groups of five interspersed by 40-s intervals during which cells were continuously paced but data were not recorded. Purified 14-3-3 (2 M) or 14-3-3 (2 M blocked with R18 peptide as described for the pull-down assay) in other experiments 2.5 g/ml of either anti-Ser16 or anti-Thr17 antibody was added to the intracellular pipette solution and dialyzed into the cell interior.

Data analysis and statisticsIn mouse myocytes, the contribution of Na+/Ca2+ exchange to the recovery of the Ca2+ transient is minimal. Li et al. (49) using mouse myocytes at 23°C estimate that SERCA is responsible for 90% of the Ca2+ transient recovery, with Na+/Ca2+ exchange contributing only 9%. In our studies at 35°C (at which active processes may be even more dominant), we found that the rapid removal of extracellular

on Decem

ber 28, 2020http://stke.sciencem

ag.org/D

ownloaded from

Page 14: BIOCHEMISTRY Copyright © 2020 14-3-3 binding creates a ... · 9/1/2020  · Center Göttingen, Robert-Koch-Straße 40, 37075 Göttingen, Germany 7Max-Planck Institute for Biophysical

Menzel et al., Sci. Signal. 13, eaaz1436 (2020) 1 September 2020

S C I E N C E S I G N A L I N G | R E S E A R C H A R T I C L E

14 of 16

Na+ (<50-ms exchange time) before a Ca2+ transient did not affect the recovery of the transient (fig. S8G). This finding suggested that at 35°C, the recovery of the Ca2+ transient was dominated by SERCA. This recovery phase could therefore be accurately described by a single exponential fit, and the time constant of this Ca transient recovery (t) could be used as an index of SERCA activity. Ca2+ transients were automatically fitted using custom software written in MATLAB (MathWorks Inc., USA). Time constants obtained for each of the five sequential Ca2+ transients in each group were aver-aged to get a single value at each time point. To minimize differ-ences caused by cell variability, data obtained in each individual cell were normalized to the maximum response amplitude. Differences between experimental groups were tested with two-way analysis of variance (ANOVA) mixed-model test, followed by a Tukey’s multi-ple comparisons test.

Computational modelingThe three-dimensional (3D) structure of 14-3-3 in complex with pSer16-PLN peptide (RRApSTIEMP—UniProt identifier P26678) was generated by homology modeling using the PrimeX software, version v5.5 [Schrodinger (50)]. Input coordinates of the template systems were retrieved from Protein Data Bank (PDB) codes 3MHR and 4DAT that contain the phosphopeptides complexed to 14-3-3. With the purpose of modeling the longest peptide sequence, amino acids from the C terminus to the phosphosite and from the phos-phosite to the N terminus were taken from 4DAT and 3MHR, respectively. The homology model was relaxed through MD simu-lations carried out by Amber18 program. The force field ff14SB (51) was used to parametrize protein and peptide, whereas the general amber force field was used for nonstandard residues. Parameters for the phosphoserine residue were retrieved from the AMBER Parameter Database (52). For MD simulation purposes, the system was solvated in a cubic box of TIP3P water molecules buffering 10 Å from the surface. Periodic boundary conditions were applied. To keep the system electrically neutral, the total charge was neutralized by the addition of Na+ counter ions. Following previous work (53, 54), the solvent and Na+ ions were energy-minimized for 1500 steps using the steepest descent (SD) algorithm and a further 3500 steps using the conjugate gradient (CG) algorithm while keeping the solute as fixed. The solvated solute was then energy-minimized for 1500 steps using the SD and 3500 steps using the CG before be-ing heated at constant volume using the Langevin thermostat from 0 to 300 K over 150 ps. A density equilibration was carried out at constant pressure (NPT ensemble) for 150 ps, before running the production of unbiased MD trajectories for 250 ns (time step was 2 fs). MD trajectories were processed by cpptraj software (55) and analyzed in terms of RMSD or residue fluctuation (RMSF) and cluster analysis. The same parameters were used to relax the apo 14-3-3 by MD simula-tions. The same procedure as described above was used to generate the 3D structure of 14-3-3 in complex with pThr17-PLN peptide (RASpTIEMP). Electrostatic surface potential was computed with APBS version 27 (56) and visualized in VMD molecular graphic soft-ware (57). The delta energy of binding was computed by the molecular mechanics Poisson-Boltzmann surface area (MM-PBSA) approach using the MMPBSA.py script in AMBER18 (58). Theoretical affinity results are presented as the delta energy of binding ± SEM (59).

SUPPLEMENTARY MATERIALSstke.sciencemag.org/cgi/content/full/13/647/eaaz1436/DC1

Fig. S1. Cardiac cytosolic fractions do not show a signal for 14-3-3–bound PLN.Fig. S2. Confocal imaging showing the colocalization of APEX2-PLN or APEX2-PLN (∆1–29) and SERCA2a in NRCMs.Fig. S3. All seven 14-3-3 isoforms bind pThr17-PLN-GST.Fig. S4. Determination of the equilibrium dissociation constant for 14-3-3 binding to pThr17-PLN.Fig. S5. MD simulation of the 14-3-3–pSer16–PLN complex and the 14-3-3–pThr17–PLN complex.Fig. S6. PKA does not phosphorylate ∆Arg14-PLN monomers.Fig. S7. 14-3-3 binding protects pSer16-T17A-PLN from fast dephosphorylation.Fig. S8. Effect of 14-3-3 and Na+-free superfusion on the kinetics of isoprenaline-induced Ca2+ transients in isolated cardiomyocytes.Table S1. Plasmids used in this study.Table S2. Antibodies used in this study.Data file S1. In vivo proximity labeling.

View/request a protocol for this paper from Bio-protocol.

REFERENCES AND NOTES 1. D. M. Bers, Cardiac excitation-contraction coupling. Nature 415, 198–205 (2002). 2. K. M. Alsina, M. Hulsurkar, S. Brandenburg, D. Kownatzki-Danger, C. Lenz, H. Urlaub,

I. Abu-Taha, M. Kamler, D. Y. Chiang, S. K. Lahiri, J. O. Reynolds, A. P. Quick, L. Scott Jr., T. A. Word, M. D. Gelves, A. J. R. Heck, N. Li, D. Dobrev, S. E. Lehnart, X. H. T. Wehrens, Loss of protein phosphatase 1 regulatory subunit PPP1R3A promotes atrial fibrillation. Circulation 140, 681–693 (2019).

3. A. G. Brittsan, A. N. Carr, A. G. Schmidt, E. G. Kranias, Maximal inhibition of SERCA2 Ca2+ affinity by phospholamban in transgenic hearts overexpressing a non-phosphorylatable form of phospholamban. J. Biol. Chem. 275, 12129–12135 (2000).

4. W. Luo, I. L. Grupp, J. Harrer, S. Ponniah, G. Grupp, J. J. Duffy, T. Doetschman, E. G. Kranias, Targeted ablation of the phospholamban gene is associated with markedly enhanced myocardial contractility and loss of beta-agonist stimulation. Circ. Res. 75, 401–409 (1994).

5. K. Haghighi, F. Kolokathis, A. O. Gramolini, J. R. Waggoner, L. Pater, R. A. Lynch, G.-C. Fan, D. Tsiapras, R. R. Parekh, G. W. Dorn II, D. H. MacLennan, D. T. Kremastinos, E. G. Kranias, A mutation in the human phospholamban gene, deleting arginine 14, results in lethal, hereditary cardiomyopathy. Proc. Natl. Acad. Sci. U.S.A. 103, 1388–1393 (2006).

6. K. Haghighi, F. Kolokathis, L. Pater, R. A. Lynch, M. Asahi, A. O. Gramolini, G.-C. Fan, D. Tsiapras, H. S. Hahn, S. Adamopoulos, S. B. Liggett, G. W. Dorn II, D. H. MacLennan, D. T. Kremastinos, E. G. Kranias, Human phospholamban null results in lethal dilated cardiomyopathy revealing a critical difference between mouse and human. J. Clin. Invest. 111, 869–876 (2003).

7. J. P. Schmitt, M. Kamisago, M. Asahi, G. H. Li, F. Ahmad, U. Mende, E. G. Kranias, D. MacLennan, J. G. Seidman, C. E. Seidman, Dilated cardiomyopathy and heart failure caused by a mutation in phospholamban. Science 299, 1410–1413 (2003).

8. E. G. Kranias, R. J. Solaro, Phosphorylation of troponin I and phospholamban during catecholamine stimulation of rabbit heart. Nature 298, 182–184 (1982).

9. A. D. Wegener, H. K. Simmerman, J. P. Lindemann, L. R. Jones, Phospholamban phosphorylation in intact ventricles. Phosphorylation of serine 16 and threonine 17 in response to -adrenergic stimulation. J. Biol. Chem. 264, 11468–11474 (1989).

10. W. Luo, G. Chu, Y. Sato, Z. Zhou, V. J. Kadambi, E. G. Kranias, Transgenic approaches to define the functional role of dual site phospholamban phosphorylation. J. Biol. Chem. 273, 4734–4739 (1998).

11. D. H. MacLennan, E. G. Kranias, Phospholamban: A crucial regulator of cardiac contractility. Nat. Rev. Mol. Cell Biol. 4, 566–577 (2003).

12. G. Chu, J. W. Lester, K. B. Young, W. Luo, J. Zhai, E. G. Kranias, A single site (Ser16) phosphorylation in phospholamban is sufficient in mediating its maximal cardiac responses to -agonists. J. Biol. Chem. 275, 38938–38943 (2000).

13. M. Gustavsson, R. Verardi, D. G. Mullen, K. R. Mote, N. J. Traaseth, T. Gopinath, G. Veglia, Allosteric regulation of SERCA by phosphorylation-mediated conformational shift of phospholamban. Proc. Natl. Acad. Sci. U.S.A. 110, 17338–17343 (2013).

14. C. F. Louis, M. Maffitt, B. Jarvis, Factors that modify the molecular size of phospholamban, the 23,000-dalton cardiac sarcoplasmic reticulum phosphoprotein. J. Biol. Chem. 257, 15182–15186 (1982).

15. D. R. Singh, M. P. Dalton, E. E. Cho, M. P. Pribadi, T. J. Zak, J. Šeflová, C. A. Makarewich, E. N. Olson, S. L. Robia, Newly discovered micropeptide regulators of SERCA form oligomers but bind to the pump as monomers. J. Mol. Biol. 431, 4429–4443 (2019).

16. K. Haghighi, P. Bidwell, E. G. Kranias, Phospholamban interactome in cardiac contractility and survival: A new vision of an old friend. J. Mol. Cell. Cardiol. 77, 160–167 (2014).

17. E. Vafiadaki, D. A. Arvanitis, D. Sanoudou, E. G. Kranias, Identification of a protein phosphatase-1/phospholamban complex that is regulated by cAMP-dependent phosphorylation. PLOS ONE 8, e80867 (2013).

18. E. C. Arakel, S. Brandenburg, K. Uchida, H. Zhang, Y.-W. Lin, T. Kohl, B. Schrul, M. S. Sulkin, I. R. Efimov, C. G. Nichols, S. E. Lehnart, B. Schwappach, Tuning the electrical properties

on Decem

ber 28, 2020http://stke.sciencem

ag.org/D

ownloaded from

Page 15: BIOCHEMISTRY Copyright © 2020 14-3-3 binding creates a ... · 9/1/2020  · Center Göttingen, Robert-Koch-Straße 40, 37075 Göttingen, Germany 7Max-Planck Institute for Biophysical

Menzel et al., Sci. Signal. 13, eaaz1436 (2020) 1 September 2020

S C I E N C E S I G N A L I N G | R E S E A R C H A R T I C L E

15 of 16

of the heart by differential trafficking of KATP ion channel complexes. J. Cell Sci. 127, 2106–2119 (2014).

19. A. Aitken, 14-3-3 proteins: A historic overview. Semin. Cancer Biol. 16, 162–172 (2006). 20. M. K. Dougherty, D. K. Morrison, Unlocking the code of 14-3-3. J. Cell Sci. 117, 1875–1884

(2004). 21. C. Mackintosh, Dynamic interactions between 14-3-3 proteins and phosphoproteins

regulate diverse cellular processes. Biochem. J. 381, 329–342 (2004). 22. K. Michelsen, H. Yuan, B. Schwappach, Hide and run. Arginine-based endoplasmic-

reticulum-sorting motifs in the assembly of heteromultimeric membrane proteins. EMBO Rep. 6, 717–722 (2005).

23. A. J. Smith, J. Daut, B. Schwappach, Membrane proteins as 14-3-3 clients in functional regulation and intracellular transport. Physiology 26, 181–191 (2011).

24. M. Kilisch, O. Lytovchenko, E. C. Arakel, D. Bertinetti, B. Schwappach, A dual phosphorylation switch controls 14-3-3-dependent cell surface expression of TASK-1. J. Cell Sci. 129, 831–842 (2016).

25. A. Lundby, M. N. Andersen, A. B. Steffensen, H. Horn, C. D. Kelstrup, C. Francavilla, L. J. Jensen, N. Schmitt, M. B. Thomsen, J. V. Olsen, In vivo phosphoproteomics analysis reveals the cardiac targets of -adrenergic receptor signaling. Sci. Signal. 6, rs11 (2013).

26. C. Petosa, S. C. Masters, L. A. Bankston, J. Pohl, B. Wang, H. Fu, R. C. Liddington, 14-3-3 binds a phosphorylated Raf peptide and an unphosphorylated peptide via its conserved amphipathic groove. J. Biol. Chem. 273, 16305–16310 (1998).

27. B. Wang, H. Yang, Y.-C. Liu, T. Jelinek, L. Zhang, E. Ruoslahti, H. Fu, Isolation of high-affinity peptide antagonists of 14-3-3 proteins by phage display. Biochemistry 38, 12499–12504 (1999).

28. S. Ghorbani, A. Fossbakk, A. Jorge-Finnigan, M. I. Flydal, J. Haavik, R. Kleppe, Regulation of tyrosine hydroxylase is preserved across different homo- and heterodimeric 14-3-3 proteins. Amino Acids 48, 1221–1229 (2016).

29. L. M. Stevers, P. J. de Vink, C. Ottmann, J. Huskens, L. Brunsveld, A thermodynamic model for multivalency in 14-3-3 protein–protein interactions. J. Am. Chem. Soc. 140, 14498–14510 (2018).

30. H. Yuan, K. Michelsen, B. Schwappach, 14-3-3 dimers probe the assembly status of multimeric membrane proteins. Curr. Biol. 13, 638–646 (2003).

31. E. W. Wilker, R. A. Grant, S. C. Artim, M. B. Yaffe, A structural basis for 14-3-3sigma functional specificity. J. Biol. Chem. 280, 18891–18898 (2005).

32. A. Ballone, F. Centorrino, M. Wolter, C. Ottmann, Protein X-ray crystallography of the 14-3-3/SOS1 complex. Data Brief 19, 1683–1687 (2018).

33. H. Fu, R. R. Subramanian, S. C. Masters, 14-3-3 proteins: Structure, function, and regulation. Annu. Rev. Pharmacol. Toxicol. 40, 617–647 (2000).

34. M. Gouw, S. Michael, H. Sámano-Sánchez, M. Kumar, A. Zeke, B. Lang, B. Bely, L. B. Chemes, N. E. Davey, Z. Deng, F. Diella, C.-M. Gürth, A.-K. Huber, S. Kleinsorg, L. S. Schlegel, N. Palopoli, K. V. Roey, B. Altenberg, A. Reményi, H. Dinkel, T. J. Gibson, The eukaryotic linear motif resource—2018 update. Nucleic Acids Res. 46, D428–D434 (2018).

35. K. Rittinger, J. Budman, J. Xu, S. Volinia, L. C. Cantley, S. J. Smerdon, S. J. Gamblin, M. B. Yaffe, Structural analysis of 14-3-3 phosphopeptide complexes identifies a dual role for the nuclear export signal of 14-3-3 in ligand binding. Mol. Cell 4, 153–166 (1999).

36. F. Centorrino, A. Ballone, M. Wolter, C. Ottmann, Biophysical and structural insight into the USP8/14-3-3 interaction. FEBS Lett. 592, 1211–1220 (2018).

37. P. Sharma, V. Ignatchenko, K. Grace, C. Ursprung, T. Kislinger, A. O. Gramolini, Endoplasmic reticulum protein targeting of phospholamban: A common role for an N-terminal di-arginine motif in ER retention? PLOS ONE 5, e11496 (2010).

38. A. Kagan, Y. F. Melman, A. Krumerman, T. V. McDonald, 14-3-3 amplifies and prolongs adrenergic stimulation of HERG K+ channel activity. EMBO J. 21, 1889–1898 (2002).

39. UniProt Consortium, UniProt: A worldwide hub of protein knowledge. Nucleic Acids Res. 47, D506–D515 (2019).

40. M. E. Anderson, J. H. Brown, D. M. Bers, CaMKII in myocardial hypertrophy and heart failure. J. Mol. Cell. Cardiol. 51, 468–473 (2011).

41. S. Sossalla, N. Fluschnik, H. Schotola, K. R. Ort, S. Neef, T. Schulte, K. Wittköpper, A. Renner, J. D. Schmitto, J. Gummert, A. El-Armouche, G. Hasenfuss, L. S. Maier, Inhibition of elevated Ca2+/calmodulin-dependent protein kinase II improves contractility in human failing myocardium. Circ. Res. 107, 1150–1161 (2010).

42. P. Beauverger, M.-L. Ozoux, G. Bégis, V. Glénat, V. Briand, M.-C. Philippo, C. Daveu, G. Tavares, S. Roy, A. Corbier, P. Briand, O. Dorchies, A.-L. Bauchet, E. Nicolai, O. Duclos, D. Tamarelle, M.-P. Pruniaux, A. J. Muslin, P. Janiak, Reversion of cardiac dysfunction by a novel orally available calcium/calmodulin-dependent protein kinase II inhibitor, RA306, in a genetic model of dilated cardiomyopathy. Cardiovasc. Res. 116, 329–338 (2019).

43. D. K. Ceholski, C. A. Trieber, C. F. Holmes, H. S. Young, Lethal, hereditary mutants of phospholamban elude phosphorylation by protein kinase A. J. Biol. Chem. 287, 26596–26605 (2012).

44. K. N. Ha, L. R. Masterson, Z. Hou, R. Verardi, N. Walsh, G. Veglia, S. L. Robia, Lethal Arg9Cys phospholamban mutation hinders Ca2+-ATPase regulation and phosphorylation by protein kinase A. Proc. Natl. Acad. Sci. U.S.A. 108, 2735–2740 (2011).

45. V. Hung, N. D. Udeshi, S. S. Lam, K. H. Loh, K. J. Cox, K. Pedram, S. A. Carr, A. Y. Ting, Spatially resolved proteomic mapping in living cells with the engineered peroxidase APEX2. Nat. Protoc. 11, 456–475 (2016).

46. I. Atanassov, H. Urlaub, Increased proteome coverage by combining PAGE and peptide isoelectric focusing: Comparative study of gel-based separation approaches. Proteomics 13, 2947–2955 (2013).

47. A. K. Snabaitis, A. Muntendorf, T. Wieland, M. Avkiran, Regulation of the extracellular signal-regulated kinase pathway in adult myocardium: Differential roles of Gq/11, Gi and G12/13 proteins in signalling by 1-adrenergic, endothelin-1 and thrombin-sensitive protease-activated receptors. Cell. Signal. 17, 655–664 (2005).

48. N. Voigt, N. Li, Q. Wang, W. Wang, A. W. Trafford, I. Abu-Taha, Q. Sun, T. Wieland, U. Ravens, S. Nattel, X. H. T. Wehrens, D. Dobrev, Enhanced sarcoplasmic reticulum Ca2+ leak and increased Na+-Ca2+ exchanger function underlie delayed afterdepolarizations in patients with chronic atrial fibrillation. Circulation 125, 2059–2070 (2012).

49. L. Li, G. Chu, E. G. Kranias, D. M. Bers, Cardiac myocyte calcium transport in phospholamban knockout mouse: Relaxation and endogenous CaMKII effects. Am. J. Physiol. 274, H1335–H1347 (1998).

50. J. A. Bell, Y. Cao, J. R. Gunn, T. Day, E. Gallicchio, Z. Zhou, R. Levy, R. Farid, PrimeX and the Schrödinger computational chemistry suite of programs. Int. Tables Crystallogr. F, 534–538 (2012).

51. J. A. Maier, C. Martinez, K. Kasavajhala, L. Wickstrom, K. E. Hauser, C. Simmerling, ff14SB: Improving the accuracy of protein side chain and backbone parameters from ff99SB. J. Chem. Theory Comput. 11, 3696–3713 (2015).

52. N. Homeyer, A. H. Horn, H. Lanig, H. Sticht, AMBER force-field parameters for phosphorylated amino acids in different protonation states: Phosphoserine, phosphothreonine, phosphotyrosine, and phosphohistidine. J. Mol. Model. 12, 281–289 (2006).

53. Y. Cau, A. Fiorillo, M. Mori, A. Ilari, M. Botta, M. Lalle, Molecular dynamics simulations and structural analysis of Giardia duodenalis 14-3-3 protein-protein interactions. J. Chem. Inf. Model. 55, 2611–2622 (2015).

54. M. Mori, A. Nucci, M. C. D. Lang, N. Humbert, C. Boudier, F. Debaene, S. Sanglier-Cianferani, M. Catala, P. Schult-Dietrich, U. Dietrich, C. Tisné, Y. Mely, M. Botta, Functional and structural characterization of 2-amino-4-phenylthiazole inhibitors of the HIV-1 nucleocapsid protein with antiviral activity. ACS Chem. Biol. 9, 1950–1955 (2014).

55. D. R. Roe, T. E. Cheatham III, PTRAJ and CPPTRAJ: Software for processing and analysis of molecular dynamics trajectory data. J. Chem. Theory Comput. 9, 3084–3095 (2013).

56. N. A. Baker, D. Sept, S. Joseph, M. J. Holst, J. A. McCammon, Electrostatics of nanosystems: Application to microtubules and the ribosome. Proc. Natl. Acad. Sci. U.S.A. 98, 10037–10041 (2001).

57. W. Humphrey, A. Dalke, K. Schulten, VMD: Visual molecular dynamics. J. Mol. Graph. 14, 33–38 (1996).

58. B. R. Miller III, T. D. McGee Jr., J. M. Swails, N. Homeyer, H. Gohlke, A. E. Roitberg, MMPBSA.py: An efficient program for end-state free energy calculations. J. Chem. Theory Comput. 8, 3314–3321 (2012).

59. S. Orchard, M. Ammari, B. Aranda, L. Breuza, L. Briganti, F. Broackes-Carter, N. H. Campbell, G. Chavali, C. Chen, N. del-Toro, M. Duesbury, M. Dumousseau, E. Galeota, U. Hinz, M. Iannuccelli, S. Jagannathan, R. Jimenez, J. Khadake, A. Lagreid, L. Licata, R. C. Lovering, B. Meldal, A. N. Melidoni, M. Milagros, D. Peluso, L. Perfetto, P. Porras, A. Raghunath, S. Ricard-Blum, B. Roechert, A. Stutz, M. Tognolli, K. van Roey, G. Cesareni, H. Hermjakob, The MIntAct project—IntAct as a common curation platform for 11 molecular interaction databases. Nucleic Acids Res. 42, D358–D363 (2014).

Acknowledgments: We thank Schwappach and Lehnart laboratory members, in particular G. Weninger, for technical help, reagents, and comments on the manuscript. Modeling used a Titan Xp GPU donated by NVIDIA Corporation. The technical assistance of B. Korff, S. Reji, and A, Wolf is acknowledged. A.-K. Ozga provided statistical advice. Funding: J.M. was funded by the DFG (Deutsche Forschungsgemeinschaft)–funded International Research Training Group (IRTG1816). Work on PLN in the Schwappach laboratory was supported by the DFG (SFB1002; A07 and under Germany’s Excellence Strategy—EXC 2067/1-390729940). D.K.-D. and S.E.L. were funded by the DFG (SFB1190; P03) and the Leducq Foundation transatlantic network CURE-PLaN, and supported under Germany’s Excellence Strategy—EXC 2067/1-390729940. S.E.L. is a principal investigator of the German Centre for Cardiovascular Research (DZHK). Mass spectrometric analyses were performed by the Core Facility Proteomics at the University Medical Center Göttingen and supported by the DFG (SFB1190; Z02). A.B., M.M., and C.O. were supported by the H2020 Marie Curie Action of the European Commission (grant number 675179). The physiological measurements in the Shattock laboratory were funded by the British Heart Foundation (RG/17/15/33106). Author contributions: J.M., M.J.S., S.E.L., and B.S. conceptualized experiments and manuscript. J.M., D.K.-D., S.T., A.B., K.U.-F., M.K., M.M., C.O., M.J.S., S.E.L., and B.S. developed and provided methodology. J.M., D.K.-D., S.T., A.B., M.M., C.O., M.J.S., S.E.L., and B.S. validated experimental data. J.M., D.K.-D., S.T., A.B., M.K., C.L., M.M., M.J.S., S.E.L., and B.S. performed formal analysis. J.M., D.K.-D., S.T., A.B., K.U.-F., M.M., M.J.S., S.E.L., and

on Decem

ber 28, 2020http://stke.sciencem

ag.org/D

ownloaded from

Page 16: BIOCHEMISTRY Copyright © 2020 14-3-3 binding creates a ... · 9/1/2020  · Center Göttingen, Robert-Koch-Straße 40, 37075 Göttingen, Germany 7Max-Planck Institute for Biophysical

Menzel et al., Sci. Signal. 13, eaaz1436 (2020) 1 September 2020

S C I E N C E S I G N A L I N G | R E S E A R C H A R T I C L E

16 of 16

B.S. drove the investigation. H.U., M.M., C.O., M.J.S., S.E.L., and B.S. provided resources. J.M., D.K.-D., S.T., A.B., C.L., M.M., M.J.S., S.E.L., and B.S. curated data. J.M., S.E.L., and B.S. wrote the original draft. J.M., D.K.-D., A.B., C.L., M.M., C.O., M.J.S., S.E.L., and B.S. wrote, reviewed, and edited the manuscript. J.M., D.K.-D., S.T., A.B., and M.M. visualized data. B.S. administrated the project. C.L., H.U., M.M., C.O., M.J.S., S.E.L., and B.S. acquired funding. Competing interests: The authors declare that they have no competing interests. Data and materials availability: The protein-protein interactions from this publication have been submitted to the IMEx (http://imexconsortium.org) consortium through IntAct (43) and assigned the identifier IM-27195. All other data needed to evaluate the conclusions in the paper are present in the paper or the Supplementary Materials.

Submitted 15 August 2019Resubmitted 9 February 2020Accepted 8 July 2020Published 1 September 202010.1126/scisignal.aaz1436

Citation: J. Menzel, D. Kownatzki-Danger, S. Tokar, A. Ballone, K. Unthan-Fechner, M. Kilisch, C. Lenz, H. Urlaub, M. Mori, C. Ottmann, M. J. Shattock, S. E. Lehnart, B. Schwappach, 14-3-3 binding creates a memory of kinase action by stabilizing the modified state of phospholamban. Sci. Signal. 13, eaaz1436 (2020).

on Decem

ber 28, 2020http://stke.sciencem

ag.org/D

ownloaded from

Page 17: BIOCHEMISTRY Copyright © 2020 14-3-3 binding creates a ... · 9/1/2020  · Center Göttingen, Robert-Koch-Straße 40, 37075 Göttingen, Germany 7Max-Planck Institute for Biophysical

phospholamban14-3-3 binding creates a memory of kinase action by stabilizing the modified state of

Henning Urlaub, Mattia Mori, Christian Ottmann, Michael J. Shattock, Stephan E. Lehnart and Blanche SchwappachJulia Menzel, Daniel Kownatzki-Danger, Sergiy Tokar, Alice Ballone, Kirsten Unthan-Fechner, Markus Kilisch, Christof Lenz,

DOI: 10.1126/scisignal.aaz1436 (647), eaaz1436.13Sci. Signal. 

distinct biological outcomes.SERCA activity than would PKA and show how different kinases acting on adjacent residues in a substrate can generatethan did PKA-phosphorylated phospholamban. These results suggest that CaMKII would have a longer-lasting effect on target sites are adjacent to each other, CaMKII-phosphorylated phospholamban interacted with 14-3-3 with higher affinityprotected phospholamban, specifically the pentameric form, from dephosphorylation. Although the PKA and CaMKII

found that the 14-3-3 family of phosphoadaptor proteinset al.effect, resulting in increased force generation. Menzel ) alleviate this inhibitory17) or CaMKII (which phosphorylates Thr16phospholamban by PKA (which phosphorylates Ser

pump SERCA. Stimuli that cause the phosphorylation of2+cardiomyocyte contraction by inhibiting the Ca signal and limits2+The membrane protein phospholamban decreases the duration of the cardiomyocyte Ca

14-3-3 remembers the kinase

ARTICLE TOOLS http://stke.sciencemag.org/content/13/647/eaaz1436

MATERIALSSUPPLEMENTARY http://stke.sciencemag.org/content/suppl/2020/08/28/13.647.eaaz1436.DC1

CONTENTRELATED

http://stke.sciencemag.org/content/sigtrans/13/659/eabf6590.fullhttp://science.sciencemag.org/content/sci/366/6461/109.fullhttp://stke.sciencemag.org/content/sigtrans/9/457/ra119.full

REFERENCES

http://stke.sciencemag.org/content/13/647/eaaz1436#BIBLThis article cites 59 articles, 23 of which you can access for free

PERMISSIONS http://www.sciencemag.org/help/reprints-and-permissions

Terms of ServiceUse of this article is subject to the

is a registered trademark of AAAS.Science SignalingYork Avenue NW, Washington, DC 20005. The title (ISSN 1937-9145) is published by the American Association for the Advancement of Science, 1200 NewScience Signaling

Science. No claim to original U.S. Government WorksCopyright © 2020 The Authors, some rights reserved; exclusive licensee American Association for the Advancement of

on Decem

ber 28, 2020http://stke.sciencem

ag.org/D

ownloaded from


Recommended